Toska Et Al. - 2017 - Sex in The Shadow of HIV A Systematic Review of P

a1111111111 a1111111111 a1111111111 a1111111111 a1111111111
PLOS ONE | https://doi.org/10.1371/journal.pone.0178106 June 5, 2017 1 / 30

OPEN ACCESS
Citation: Toska E, Pantelic M, Meinck F, Keck K, Haghighat R, Cluver L (2017) Sex in the shadow of HIV: A systematic review
of prevalence, risk factors, and interventions to reduce sexual risk- taking among HIV-positive adolescents and youth in
sub-Saharan Africa. PLoS ONE 12(6): e0178106. https://doi.org/10.1371/journal.pone.0178106
Editor: Omar Sued, ARGENTINA
Received: February 2, 2017
Accepted: May 6, 2017
Published: June 5, 2017
Copyright: © 2017 Toska et al. This is an open access article distributed under the terms of the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source
are credited.
Data Availability Statement: All relevant data are within the paper and its Supporting Information files.
Funding: ET, MP, and KK were supported by the Evidence for HIV Prevention in Southern Africa
[MM/EHPSA/UCT/05150014]. Additional support for MP was provided by the joint Green Templeton- Clarendon Scholarship
fund. RH was supported by the Economic and Social Research Council (ESRC) Doctoral Training Centre Studentship and the
Clarendon-Jesus College Old Members’ Fund. FM
RESEARCH ARTICLE Sex in the shadow of HIV: A

systematic review of prevalence, risk factors, and interventions
to reduce sexual risk-taking among HIV- positive adolescents
and youth in sub- Saharan Africa
Elona Toska1,2*, Marija Pantelic1☯, Franziska Meinck1,3☯, Katharina Keck1,4☯, Roxanna Haghighat1‡,
Lucie Cluver1,5‡
1 Department of Social Policy and Intervention, University of Oxford, Oxford, United Kingdom, 2 AIDS and Society
Research Unit, Centre for Social Science Research, University of Cape Town, Cape Town, South Africa, 3
OPTENTIA, School of Behavioural Sciences, North-West University, Vanderbeijlpark, South Africa, 4 Oxford Policy
Management, Johannesburg, South Africa, 5 Department of Psychiatry and Mental Health, University of Cape Town,
Cape Town, South Africa
☯ These authors contributed equally to this work. ‡ These authors also contributed equally to this work. 4
elona.toska@gmail.com
Abstract
Background
Evidence on sexual risk-taking among HIV-positive adolescents and youth in sub-Saharan Africa is urgently needed.
This systematic review synthesizes the extant research on prevalence, factors associated with, and interventions to
reduce sexual risk-taking among HIV- positive adolescents and youth in sub-Saharan Africa.
Methods
Studies were located through electronic databases, grey literature, reference harvesting, and contact with
researchers. Preferred Reporting Items for Systematic Reviews and Meta- Analyses guidelines were followed.
Quantitative studies that reported on HIV-positive participants (10–24 year olds), included data on at least one of
eight outcomes (early sexual debut, inconsistent condom use, older partner, transactional sex, multiple sexual
partners, sex while intoxicated, sexually transmitted infections, and pregnancy), and were conducted in sub-Saharan
Africa were included. Two authors piloted all processes, screened studies, extracted data independently, and
resolved any discrepancies. Due to variance in reported rates and factors associated with sexual risk-taking,
meta-analyses were not conducted.
Results
610 potentially relevant titles/abstracts resulted in the full text review of 251 records. Forty- two records (n = 35
studies) reported one or multiple sexual practices for 13,536 HIV-positive adolescents/youth from 13 sub-Saharan
African countries. Seventeen cross-sectional

Sexual risk-taking among HIV-positive adolescents in sub-Saharan Africa: A systematic review
and LC were supported by the European Research
studies reported on individual, relationship, family, structural, and HIV-related
factors asso- Council (ERC) under the European Union’s Seventh Framework Program [FP7/2007-2013]/ERC grant agreement n ̊
313421, University of Oxford’s ESRC
ciated with sexual risk-taking. However, the majority of the findings were inconsistent across studies, and most
studies scored <50% in the quality checklist. Living with a partner, living
Impact Acceleration Account (grant 1311-KEA-004
alone, gender-based violence, food insecurity, and employment were
correlated with & 1609-GCRF-227), and the Philip Leverhulme
increased sexual risk-taking, while knowledge of own HIV-positive status and
accessing Trust [PLP-2014-095 AQ6]. The funders had no role in study design, data collection and analysis, decision to publish,
or preparation of the
HIV support groups were associated with reduced sexual risk-taking. Of the four intervention studies (three RCTs),
three evaluated group-based interventions, and one evaluated an
manuscript.
individual-focused combination intervention. Three of the interventions were effective at
Competing interests: The authors have declared
reducing sexual risk-taking, with one reporting no difference between the intervention and
that no competing interests exist.
control groups.
Conclusion
Sexual risk-taking among HIV-positive adolescents and youth is high, with inconclusive evidence on potential
determinants. Few known studies test secondary HIV-prevention interventions for HIV-positive youth. Effective and
feasible low-cost interventions to reduce risk are urgently needed for this group.
Introduction
With increased access to antiretroviral treatment in sub-Saharan Africa, the number of children vertically infected
with HIV who survive to adolescence has risen [1,2]. Coupled with sus- tained high HIV-incidence among youth in
the region, this has resulted in nearly 1.7 million HIV-positive adolescents (10–19 years old) in sub-Saharan Africa,
with girls representing nearly two-thirds of this total [3–5]. Despite global reductions in HIV prevalence, rates of
new HIV infections remain the highest among 15–24 year old youth in sub-Saharan Africa [6]. As their numbers
continue to grow, adolescents and youth living with HIV are an essential group for secondary HIV prevention
efforts [7].
HIV-positive adolescents and youth are at risk of passing on the virus to their sexual partners and children [8,9].
They are additionally vulnerable to potential re-infection by HIV and more vulnerable to other sexually transmitted
infections (STIs) compared to their HIV-negative peers [10]. Adolescents are more likely than adults or younger
children to adhere poorly to their medication [11–13] and in particular to treatment regimens to prevent mother-to-
child-transmission [14]. Low adherence and retention in care rates are strongly associated with resistance to
available antiretroviral therapies, including second-line treatment when available [15,16]. With limited access to
second and third-line antiretroviral treatment, HIV- positive adolescents risk running out of treatment options or
infecting others with resistant strains of the virus. In addition, HIV-positive adolescents experience a range of
vulnerabilities that reduce the efficacy of generalised prevention programmes, including cognitive and mental health
issues [17,18], family-related challenges [19,20], and material deprivation [21,22]. Ado- lescents living with HIV in
sub-Saharan Africa are particularly vulnerable to these risks due to poor access to healthcare services such as family
planning, HIV testing, and treatment [23–27].
A small number of studies on adolescents living with HIV in sub-Saharan Africa report high rates of unprotected
sex [28–30]; however, little is known about rates of other high-risk practices, such as transactional sex, sex with
older partners, and multiple concurrent sexual partners [31]. In the general adolescent population, these high-risk
sexual practices have been

associated with higher odds of becoming infected with HIV [32]. Though the evidence on different high-risk
sexual practices among HIV-positive adolescents is nascent, understanding factors associated with sexual risk-taking
is crucial for intervention development.
Although some interventions to reduce sexual risk behaviours have been conducted among HIV-positive
adolescents in the United States [33–37], there is a dearth of research and interventions on secondary prevention
among HIV-positive adolescents in the developing world [38]. A 2010 WHO review of behavioural interventions
for HIV positive prevention in middle and lower-income countries found 19 studies, none of which focused on
young people [39]. A recent review of sexual and reproductive health and rights interventions for youth living with
HIV in sub-Saharan Africa located six small-scale interventions [38], only three of which quantitatively measured
change in a sexual risk behaviour [40–42].
To fill the evidence gap in effective interventions for this vulnerable population, further research is needed to
elucidate HIV-positive adolescent sexual and reproductive health needs. This includes a better understanding of the
epidemiology of sexual risk-taking as well as hypothesized models of sexual health decision-making among
HIV-positive adolescents and youth [43]. This systematic review synthesizes existing evidence of sexual risk-taking
among HIV-positive adolescents and youth in sub-Saharan Africa (10–24 years old) on: 1) prevalence 2) factors
associated with of risk taking, and 3) interventions.
Methods
This review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)
guidelines [44]. The scope of this review (Table 1) is to assess the state of the evidence for three research
questions:
1. What is the prevalence of sexual risk-taking among HIV-positive adolescents and youth in
sub-Saharan Africa?
2. What factors (correlates, risk factors, or predictors) are associated with sexual risk-taking
among HIV-positive adolescents and youth in sub-Saharan Africa?
3. Which interventions, aimed at reducing sexual risk-taking among HIV-positive adolescents
and youth in sub-Saharan Africa, have been tested, and how effective were they?
Inclusion criteria applied consisted of study population, design, sampling strategy, outcome measures,
population type, and language (S1 Table). To document outcome prevalence and factors associated with the
outcomes, cross-sectional surveys and longitudinal prospective
Table 1. Scope of systematic review.
Population Adolescents and Youth living with HIV
Age range: 10–24 years old Outcome Individual risk behaviours: early sexual debut, unprotected sex
(inconsistent condom
use/ contraception use), having an older partner, transactional sex, having multiple sexual partners, sex drunk or on
drugs, sexually transmitted infections, and unintended adolescent pregnancies. OR Composite risk behaviours
consisting of any of the above behaviours combined. Geographic location
PLOS ONE | https://doi.org/10.1371/journal.pone.0178106 June 5, 2017 3 / 30 sub-Saharan Africa
Study Design Randomised controlled trials (individual or cluster), Quasi-experimental studies
including quasi-randomized trials, controlled before-after studies, pre- and post-test studies, longitudinal cohort
studies, cross-sectional studies
https://doi.org/10.1371/journal.pone.0178106.t001

cohort studies were included. Although Randomised Controlled Trials (RCT) provide the strongest form of evidence
about intervention impact, due to the small number of RCTs identified in preliminary searches, this review also
included studies with less rigorous designs: pre-post intervention comparisons and post-intervention comparisons
with ‘control’ populations. Studies measuring at least one of eight high-risk sexual practices either as a primary or
secondary outcome were included. High-risk sexual practices included early sexual debut, unprotected sex
(inconsistent condom use/ contraception use), having an older partner, transactional sex, having multiple sexual
partners, sex whilst intoxicated, sexually transmitted infections, and unwanted adolescent pregnancies, or a
composite measure of two of these outcomes–as defined by each study. Reports in English and French were
reviewed to allow for publications from Western and Central Africa.
Exclusion criteria: Studies of special populations such as sex workers, men who have sex with men, truck drivers,
male factory workers, were excluded for three main reasons. First, the focus of the review was adolescents and
youth living with HIV in Sub-Saharan Africa, not key populations. Second, these key populations at high risk of
HIV-infection are likely to report high rates of sexual risk-taking which follow patterns not similar to those among
adolescents living in HIV-endemic communities, and thus may have biased any conclusions reached by this review.
Third, the majority of the studies of key populations focused on HIV-negative populations including only small
sub-samples of HIV-positive participants.
Search Strategy: In September-November 2015, the first author searched the online databases of
PsycARTICLES, Embase, Global Health, MEDLINE, and PsycINFO, PubMed, CINAHL, ProQuest, and WHO
Afro Library, the Cochrane and Campbell databases and the PROSPERO register of systematic reviews. The first
author also searched International AIDS Society conference abstracts and presentations, as well as websites of major
international and regional organisations, such as the World Health Organization (WHO), Joint UN Program for
HIV/AIDS (UNAIDS), the UN Children’s Fund (UNICEF), United States Agency for Interna- tional Development
(USAID), UN Family Planning Agency (UNFPA), International Planned Parenthood Federation (IPPF), and
Population Council. Key search terms for sample population (children, adolescents, teenagers and youth), all
high-risk sexual practices, location (sub- Saharan Africa) and timeline were included (S2–S4 Tables). All searches
were conducted within the publication date limits of 1983 or the closest date limit available, reflecting the time since
HIV has been diagnosed in adolescents and youth. Search terms were adapted to include the requirements of
different databases and were included in the systematic review protocol: (PROSPERO registration number
CRD42015025871).
Screening: The screening process followed the Cochrane Collaboration Handbook guidelines [45]. Following
merging and de-duplication, two authors reviewed titles and abstracts for relevance. When available, full-text
documents were retrieved and checked for eligibility against inclusion and exclusion criteria (S1 Table), and a set of
pre-agreed screening questions (S5 Table). Email requests for clarifications, unpublished data, and data from
published studies were sent to researchers working on sexual risk-taking of HIV-positive adolescents and youth.
Recent guidance on systematic reviews suggests that there is a potential bias from including studies with very small
samples in systematic reviews [46]. To minimise this bias, when studies reported !50 HIV-positive adolescents and
youth, but age-disaggregated data was not available in the included reports, authors were contacted for
age-disaggregated data for 10–24 year old HIV-positive participants. If additional data were provided, the studies
were included in the review. Reference lists of the included studies and of other relevant reviews were screened for
further eligible titles.
Data extraction: Data was extracted from full-text records by the first author (ET) using a pre-piloted data
extraction form (S1 File). A second independent reviewer checked the data

extraction for each included study (MP/FM/RH) and any discrepancies were resolved through discussion. Records
reporting analyses from the same dataset were checked for data duplication, with the largest sample taken if
multiple reports were available for the same outcome measure. For longitudinal studies reporting a change in an
outcome of interest, baseline values of the reported outcome were extracted as prevalence. If data was not reported
for HIV-positive adolescents or youth specifically but authors provided the raw data, the prevalence for sexual
risk-taking was calculated for HIV-positive adolescent or youth, via frequencies in SPSS. In such instances, the
same definition of the sexual risk outcome as the primary study was used. For example, Viegas and colleagues
reported rates of early sexual debut defined as ‘before the age of 18’ for a sero-assorted sample [47]. Using a dataset
shared by the research team, this review’s first author computed the prevalence of ‘sexual debut before 18 years old’
for HIV- positive youth. Where relevant, the prevalence of risky sexual practice was computed based on the
prevalence of related safe sexual practices reported. For example, if a study reported condom use at last intercourse
as 40%, the rate of unprotected sex at last intercourse was computed as 60%. Both reported and computed
prevalence of inconsistent condom use/ unprotected sex are reported.
Risk of bias across studies was assessed using a Study Quality Checklist and risk assessment form (S2 File). The
form drew from guidance on assessing systematic bias from the Cochrane Handbook for experimental designs
(randomised controlled trials (RCTs), non-randomised controlled trials and pre- and post-test experimental design)
[45], and the Cambridge Quality Checklist for systematic reviews of risk factors [48]. The checklist was adapted in
line with a systematic review of internalised stigma among people living with HIV [49]. Adaptation included
assessing sampling strategies at two levels: facility/ community and individual level, and assessing each individual
association between potential factors and the outcome of interest. For each potential determinant, each
outcome-determinant relationship was scored as a percentage of the total score possible from the Study Quality
Checklist (SQC). SQC scores for each outcome-predictor relationship are reported in S6 Table.
Data synthesis: Given the diversity of primary studies and outcomes measured, and the cross-sectional nature of
the majority of the included studies, a meta-analysis was not conducted, in order to avoid potentially misleading
conclusions [48]. To reflect the diversity of reported prevalence rates, data was reported as the range of reported
values for studies using the same definitions for each outcome.
Results
The results of this review are reported in five sections: (1) characteristics of included studies, (2) quality assessment
of included studies, (3) prevalence of sexual risk outcomes, (4) factors associated with sexual risk-taking, and (5)
interventions addressing sexual risk-taking.
Characteristics of included studies

Fig 1 shows the PRISMA flow diagram for included studies. The PRISMA checklist is available as supporting
information. Results from different database searches were merged, resulting in 3,314 records. Grey literature
searches and hand-searches of references of included studies resulted in an additional 61 records. After
de-duplication, two authors (ET/KK) reviewed 610 titles and abstracts and the full text documents for 251 results. A
total of 42 records were included in this systematic review, which reported data from k = 35 studies (Table 2).
Study design. The 35 included studies reported data from N = 13,536 HIV-positive adolescents and youth living
in 13 countries. Four studies described interventions evaluated

Fig 1. PRISMA flow diagram for identifying included studies.
https://doi.org/10.1371/journal.pone.0178106.g001

Table 2. Summary of included studies.
First author, year Location Study
Target population HIV+ adolescents/ design1
yout
h Ankunda 2011 [77]; Ankunda 2016 [50]
Uganda CS FP; TR 15–24 year old HIV+ n = 425 (2011) n = 335
(20
16) Bakeera-Kitaka 2008 [58] Uganda CS FP; PS 15–24 year old HIV+ n = 75 Banura 2008 [57] Uganda CS FP; CS
12–24 year old HIV+ and
HIV-
n = 82
Baryamutuma 2010 [59] Uganda CS FP; CS 13–19 year old HIV+ n = 386 Beyeza-Kasheysa 2011 [60] Uganda CS
(PCS) FP; CS 15–24 year old HIV+ &
HIV-
n = 276 (2009) n = 206 (2011) Birungi 2009 [51]; Birungi 2009 [74]; Obare 2010 [118]
Uganda CS FP-CP; CS 15–19 year old HIV+ and
n = 732 status unknown) Birungi 2011 [61] Kenya CS FP; PS 15–19 year old HIV
+ females
n = 757
Cataldo 2012 [28] Malawi, Mozambique, Zambia,
Zimbabwe
CS NR; CS 10–19 year old HIV+ n = 1,703
Gavin 2006 [68] Zimbabwe CS CRS; TR 15–19 year old n = 192 Gray 1998 [69] Uganda CS (RCT) CRS; TR 15–49
year old females n = 361 Heffron 2010 [70] Eastern Africa: Kenya, Rwanda,
Tanzania, Uganda Southern Africa: Botswana, South Africa, Zambia
CS CRS; TR 18–45 year old females n = 523
Hendriksen 2007 [71]; Steffenson 2011 [83]
South Africa CS CRS; TR 15–24 year old n = 1,235 (2007);
n =
1,091 (2011) Hoffman 2008 [62] Malawi CS (PCS) FP; PS 18+ year old HIV+ females n = 90 Holub 2010 [63]
Democratic Republic of Congo CS FP; PS 14–24 year old HIV+ n = 103 Kaggwa 2012 [125] Uganda CS FP; NR
16–24 year old HIV+ n = 453 Katusiime 2012 [56] Uganda CS FP; PS 15–24 year old HIV+ n = 148 Kembo 2012 [72]
Zimbabwe CS CRS; NR 15–24 year old HIV+ and
HIV-
n = 477
Lightfoot 2007 [41] Uganda RCT CNR; CS 14–21 year old HIV+ n = 100 Malaju 2013 [67] Ethiopia CS FP; NR 15–24
year old HIV+ and
HIV-
n = 104
Mbalinda 2015 [52]; Mbalinda 2015 [114]
Uganda CS FP-FRS; PS 10–19 year old HIV+ n = 624
Mhalu 2013 [76] Tanzania CS FRS; TR 15–24 year old HIV+ n = 232 Morris 2012 [31] Cameroon CS FRS-CP; TR
12–26 year old HIV+, HIV-,
status unknown
n = 114
Muyindike 2012 [65] Uganda CS FP; NR 18–49 year old women
(n = 826)
n = 211
Nhamo 2013 [87]; Nhamo 2014 [78]
Zimbabwe RCT NR; NR 16–19 year old HIV+ n = 710
No ̈stlinger 2015 [55] Uganda CS (PCS) FC-CC; PS 13–17 year old HIV+ N = 532 Obare 2010 [53] Kenya CS FP;
PS 15–19 year old HIV+ n = 606 Pascoe 2015 [73] Zimbabwe CS CRS; TR 18–22 year old HIV+ and
HIV-
n = 199
Santelli 2013 [32] Uganda CS (PCS) CRS; TR 15–24 year old HIV+ and
HIV-
n = 204
Senyonyi 2012 [42] Uganda RCT FP; CS 12–18 year old HIV+ n = 115 Shisana 2014 [29] South African CS CRS; TR
All ages HIV+ and HIV- n = 443 Snyder 2014 [40] South Africa PPT FC-CC; CS 16–24 year old HIV+ n = 65 Test
2012 [75] Rwanda CS FC-CC; CS 16–24 year old HIV+ n = 107 Toska 2015 [54] South Africa CS FRS-CP; TR 10–19
year old HIV+ n = 858
(Continued)
Sampling strategy2

Table 2. (Continued)
First author, year Location Study
Target population HIV+ adolescents/ design1
yout
h Viegas 2015 [47] Mozambique CS FC; PS 18–24 year old HIV+ and
HIV-
n = 85
Wanyenze 2011 [66] Uganda CS FP; CS 15–49 year old HIV+ n = 159
1 CS cross-sectional study, PCS prospective cohort survey, PPT pre and post-test; RCT randomised controlled trial.
2 Sampling strategy documented at two levels: (1) cluster: facility random and/ or stratified (FRS); facility purposeful
(FP), community random and/or stratified (CRS), community purposeful (CP), mixed–see symbols (2) individual: total/
random (TR), purposeful (PS), convenience (CS); NR–not reported.
Sampling strategy2
through RCTs (k = 3) or pre- and post-test experimental design (k = 1), and the remaining k = 31 reported on
cross-sectional data.
Participant characteristics. Participants were mostly female (k = 35 studies: 47%-100%), vertically infected (k = 9
studies: 43%-100%), and on ART (k = 9 studies: 0%-88%). Of the 17 studies that reported whether HIV-positive
adolescents and youth knew their status, the majority recruited only adolescents who knew their status (k = 13). In
the six studies which recorded disclosure of HIV status to others, just under half of HIV-positive adolescents had
shared their HIV-positive status with their partners (31%-74%) [50–55].
Outcome measures. Thirty-three studies assessed sexual practices of HIV-positive adolescents as the primary
outcome, and two reported them as secondary outcomes. The outcomes measured varied in terms of the recall
periods of measurement and exact definitions (Table 3). Nine studies reported on only one sexual practice, while the
rest reported on two or more sexual practices. The most common definitions for each reported outcome were: (1)
sex before 15 years old (k = 3), (2) sex before 18 years old (k = 3), (3) inconsistent/no condom use at last sexual
encounter (k = 14), (4) current use of modern contraception (k = 6), (5) having an older partner at first sexual
intercourse (k = 4), (6) having ever had transactional sex (k = 3), (7) multiple sexual partners in the past 12 months
(k = 5), (8) sex while intoxicated (k = 2), (9) ever having had an STI or STI symptoms (k = 5), and (10) ever been
pregnant (k = 7). All outcomes were based on self-reports, except for two studies reporting results of STI tests for
Hepatitis B (HBV) [56] and Human Papillomavirus (HPV) [57]. One study reported on a composite sexual
risk-taking score [42]. In addition to the above high-risk sexual practices, three studies reported on risk-exposure
sexual outcomes such as forced sex, non-consensual first sex, gender-based violence, as their main outcomes.
These outcomes were beyond the scope of the ini- tial study protocol, therefore information on sexual risk-exposure
outcomes is provided in S7 Table.
Quality assessment of included studies

Of the seventeen studies that reported on potential risk factors or intervention effects, most scored below 50% in the
Study Quality Checklist (k = 14, range 10%-75%, S6 Table). The reasons for the low scores included study design
and analyses, sampling strategies, response/ retention rates, and sample size, which are discussed in this section.
Study design. Of the included studies sixteen were cross-sectional, three were prospective cohorts, one was an
experimental pre- and post-test, and seven were RCTs. The three prospective cohort studies did not report analyses
of change, nor did they assess longitudinal predictors of sexual risk-taking; hence, only relevant baseline data was
extracted on prevalence and potential associated factors. Data from four RCTs reported only on prevalence or
factors associated with the outcomes of interest using cross-sectional data from the baseline of the study.

Table 3. Prevalence rates of sexual risk-taking among HIV-positive adolescents and youth.
Outcome Definition Studies Rates (% or M) Gender or age disaggregation
(NR–not reported)
Actual reported outcome (actual value reported in the study)
Early sexual debut Not defined Ankunda 2011 [77] 35.1% NR Abstinence/ never having had sex (64.9%
Baryamutuma 2010 [59]
39.9% NR Early sexual initiation
Birungi 2009 [51] 33% 31% F; 37%M Ever had sex (all under 19 years old)
Cataldo 2012 [28] Mal: 10%; 38% NR Ever had sex (all under 19 years old)
Moz: 11%; 65% Moz: 10–14 y.o. (4%F; 17% M)
15–19 y.o. (69% F; 49% M)
Zam: 5%; 29% NR
Zim: 3%; 16% NR
Lightfoot 2007 [41] 40% NR Ever had sex (all under 19 years old)
Obare 2010 [118] 84% 88% F; 73% M Ever had sex (all under 19 years old)
Toska 2015 [54] 15.1% 19.2% F; 10.7% M Ever had sex (all under 19 years old)
Age of sexual debut Ankunda 2016 [50] 16.9 (2.62) 17.1 (2.41) F; 15.4 (3.85) M Mean age (SD) of sexual debut
Gavin 2006 [68] 16.4 (0.14) NR Mean age (SD) of sexual debut
Morris 2012 [31] 17.2 (2.3) NR Mean age (SD) of sexual debut
No ̈stlinger 2015 [55] 13 (11–14) NR Median age (IQR) of sexual debut
Test 2012 [75] 17 (15–18) 17 (15–18) F; 16 (15–17.5) M Median age (IQR) at consensual sexual debut
<15 years old Kembo 2012 [72] 28.1% NR
Mhalu 2013 [76] 51.5% 68% F; 84.8% M
Shisana 2014 [29] 7.4% 4.4% F; 19.1% M
15 years old Bakeera-Kitaka
2008 [58]
42.1% NR
Holub 2010 [63] 48.5% 45% F; 65% M
<18 years old Malaju 2013 [67] 63.6% NR
Pascoe 2015 [73] 36.2% NR
Viegas 2011 [47] 77.6% 78.6% F; 70% M
19 years old Kembo 2012 [72] 87.0% NR Sex between 15–19 58.9% & sex before 15 28.1%
Inconsistent condom use/ Unprotected sex
Not clearly defined Ankunda 2011 [77] 61.7% NR Condom use (38.3%)
Baryamutuma 2010 [59]
56.3% NR Consistent condom use (43.7%)
Beyeza-Kashesya 2011 [60]
28.2% NR No contraception at all (28.2%)
Heffron 2010 [70] 76% NR Consistent condom use (24%)
Mhalu 2013 [76] 58.6% NR Unprotected sex
Pascoe 2015 [73] 58.6% NR Used condoms at every sexual act (41.4%)—timeline not defined
82.4% NR Sometimes/ never used condoms with any partners (82.4%)
Never Beyeza-Kashesya
2011 [60]
39% NR Never used condoms (39%)
Morris 2012 [31] 11.9% NR Ever used condoms (88.12%)
No ̈stlinger 2015 [55] 43.5% NR Ever used a condom (56.5%)
First sex Kembo 2012 [72] 77.4% NR Condom use (22.6%)
No ̈stlinger 2015 [55] 75% NR Condom use at first sex (25%)
Toska 2015 [54] 23.6% 21.1% F;; 16.2% M Unprotected sex (25.8%, 23.6% F, 16.2% M)
Last sex Ankunda 2016 [50] 46.7% NR Consistent condom use (53.3%)
Gavin 2006 [68] 80.3% All female Condom use at last sex (19.7%)
Hendriksen 2007 [71]
37.9% 51.3% F; 55% M Unprotected sex
Holub 2010 [63] 18.5% NR Unprotected sex
Kaggwa 2012 [125] 49.5% NR No condom use at last sex (49.5%)
Kembo 2012 [72] 82.7% NR Condom use (17.3%)
Mbalinda 2015 [114] 43.7% NR Unprotected sex (43.7%)
Mhalu 2013 [76] 55.3% NR Condom use (44.7%)
Morris 2012 [31] 45% NR Condom use (55%)
Pascoe 2015 [73] 58.8% NR Unprotected sex
Shisana 2014 [29] 39.1% NR Condom use at last sex (60.9%, 62.2% F, 55.5% M)
Snyder 2014 [40] 29% 37.8% F; 44.5% M Condom use at baseline (pre-intervention 71%)
Toska 2015 [54] 28.9% 31.9% F; 13.5% M Unprotected sex at last sex (28.95%, 31.9% F, 13.5% M)
Viegas 2015 [47] 37.6% 38.7% F; 30% M Condom use (62.4%, 61.3% F, 70% M)
Last 3 months Lightfoot 2007 [41] 87.5% NR Always use condoms at baseline for full intervention sample (12.5%)
Last 6 months Ankunda 2016 [50] 53.6% NR Consistent condom use (46.4%)
Mbalinda 2015 [52] 77.5% NR Use at every sexual act (22.5%)
Test 2012 [75] 56% 57% F; 50% M Inconsistent condom use (56%, 57% F, 50% M)
Last 12 months Santelli 2013 [32] 88.7% 92.6% F; 78.6% M Never or inconsistent condom use (88.7%, 92.6% F,
78.6% M)
Current Birungi 2009 [51] 48% 41% F; 52% M Condom use: sometimes or rarely
Obare 2010 [118] 60.7% 61.8% F; 59.2% M Current using condoms (39.3%, 38.2% F, 40.8% M)
Morris 2012 [31] 69.3% NR Condom use: sometimes or never (69.3%) vs. always (30.7%)
Obare 2010 [118] 14% 16% F; 4% M Condom use (86%, 84% F, 96% M)
(Continued)

(NR–not reported)
Contraception Ever any method Beyeza-Kasheysa
2009 [60]
33% 33% Ever used any contraception (n = 345)
Birungi 2009 [74] 49.6% 43.8% F; 58.2% M Any contraception used in past/current relationships
Obare 2010 [118] 72% 72% F; 70% M Ever used contraception
Wanyenze 2011 [66]
78.7% NR Any family planning method (including traditional)
Ever pill/ injectable Obare 2010 [118] 12.5% All female Pill OR injectable among sexually active females
First sex Birungi 2009 [51] 36.4% 37.5% F; 34.7% M Used a method to prevent HIV infection/ reinfection at first sex
Obare 2010 [118] 14% 15% F; 12% M Any method
Current modern Beyeza-Kashesya
2011 [60]
33.9% NR Contraception use at baseline
Obare 2010 [118] 42.6% 41% F; 44.9% M Current modern contraception
Hoffman 2008 [62] 36.7% All female At baseline
Muyindike 2012 [65] 29.9% All female Current use at enrolment in study
Obare 2010 [118] 66% 65% F; 68% M Any method
Wanyenze 2011 [66]
63.8% NR Any modern method of family planning
Wanyenze 2011 [66]
25.5% NR Any effective method of family planning (excluding condoms)
Current pill/ injectable Beyeza-Kashesya
2011 [60]
21% All female Hormonal contraception
Oral contraceptives Heffron 2010 [70] 4.0% All female At baseline
Obare 2010 [118] 6% 6% F; 3% M At baseline
Injectable Heffron 2010 [70] 14.7% All female At baseline
Implants Heffron 2010 [70] 0.6% All female At baseline
Hysterectomy Heffron 2010 [70] 0.6% All female At baseline
Post-partum contraception
Birungi 2011 [61] 61% All female After pregnancies had ended
Contraception uptake Beyeza-Kashesya
2011 [60]
28.4% NR Started using any methods of contraception in past 12 months
Long-term contraception use
17.8% NR At baseline, 6 and 12 months follow up
Heffron 2010 [70] 26.3% All female % quarterly visits reporting contraception use during 24-month follow-up
Discontinued contraception
Beyeza-Kashesya
23.6% NR Stopped using contraception during 12 month follow-up period 2011 [60]
Dual method use Beyeza-Kashesya
2009 [60]
5% NR Unclear definition
Older sexual partner First sexual partner Gavin 2006 [68] 6.6 (0.87) All female Age difference to partner–mean (SD)
Morris 2012 [31] 65.7% NR 6+ years older
Obare 2010 [118] 52% 61% F; 20% M Older
Obare 2010 [118] 4% 4% F; 3% M Much older
Test 2012 [75] 66% 75% M; 17% M >5 years older
Test 2012 [75] 37% 41% F; 17% M >10 years older
Any Pascoe 2015 [73] 28.1% NR 6–10 years old
Pascoe 2015 [73] 18.1% NR 11+ years older
Last sexual partner Gavin 2006 [68] 5.0 (0.67) All female Age difference between participant and partner–mean (SD)
Current sexual partner
Obare 2010 [118] 56% 68% F; 10% M Older partner
Obare 2010 [118] 9% 11% F; 1% M Much older
Shisana 2014 [29] 35.4% 40.6% F; 12.7% M Other options: 5+ years younger, less than 5 years difference
Transactional sex Ever Holub 2010 [63] 22.3% 23.3% F; 17.6% M Ever received money for sex
Pascoe 2015 [73] 22.6% NR Had sex with partner for material/other support
Test 2012 [75] 20.6% 66% F; 17% M Ever had sex for money
Last partner Gavin 2006 [68] 37.9% NR Received goods or money for sex with the last partner
Not clearly defined Nhamo 2014 [78] 60% All female Not clear, at baseline of RCT
(Continued)

(NR–not reported)
Multiple sexual partners Number of lifetime
partners
Beyeza-Kashesya
3 NR Unclear if mean or median, IQR (1–4) 2011 [60]
Test 2012 [75] 2.5 (1–5) 3 (1–6) F; 3 (2–4.8) M Median (IQR)
Lifetime Gavin 2006 [68] 15.7% All female Multiple lifetime sexual partners
Morris 2012 [31] 81.2% NR 2 or more lifetime sexual partners
Pascoe 2015 [73] 26.1% NR 2 or more lifetime sexual partners
Viegas 2015 [47] 88.2% 86.7% F; 100% M More than 1 lifetime sexual partner
Last 6 months Mbalinda 2015 [114] 16% NR More than 1 partner
Nhamo 2014 [78] 6% All female Multiple sexual partners at baseline of RCT
Last 12 months Steffenson 2011
[83]
11.5% 7.1% F; 30% M Concurrent partnerships
Holub 2010 [63] 7.7% 9% F; 0% M 2 or more partners
Kembo 2012 [72] 4.6% NR 2 or more sexual partners
Santelli 2013 [32] 18.1% 12.2% F 33.9% M 2 sexual partners
Santelli 2013 [32] 10.8% 2.0% F; 33.9% M 3 or more sexual partners
Santelli 2013 [32] 9.8% 2.0% F; 30.4% M 2 or more sexual partners from outside the community
Shisana 2014 [29] 15.4% 11.8% F; 29.5% M 2 or more sexual partners
Current Ankunda 2016 [50] 30.4% NR At time of interview
18% NR Current polygamous relationships
Mhalu 2013 [76] 14.5% 15.9% F; 10.6% M Compared to those with 0–1 sexual partners
Santelli 2013 [32] 13.2% 4.7% F; 35.7% M Concurrent partnerships at interview
Not clearly defined Bakeera-Kitaka
2008 [58]
36.8% NR 2 or more sexual partners
Sex intoxicated Sex after alcohol Shisana 2014 [29] 5.5% 5.8% F; 4.2% M Drank alcohol before sex with most recent
partner
Shisana 2014 [29] 3.7% 4.9% F; 1.7% M Drank alcohol before sex with the second most recent partner
Test 2012 [75] 29% NR Drank alcohol up to 6 hours before sex
STIs Ever had STIs Baryamutuma 2010
[59]
17.6% NR Self-reported
Gavin 2006 [68] 21.9% All female Self-reported symptoms
Mbalinda 2015 [52] 13.1% NR Ever had STI treatment (self-reported)
Mbalinda 2015 [52] 14.8% NR Ever had STI symptoms: genital sores
Mbalinda 2015 [52] 27.7% NR Ever had STI symptoms: genital itching
Mbalinda 2015 [52] 10.9% NR Ever had STI symptoms: genital discharge
Mbalinda 2015 [52] 16.8% NR Ever had STI symptoms: lower abdominal pain
Pascoe 2015 [73] 46.2% NR Ever had STI symptoms
Viegas 2015 [47] 36.5% 34.7% F; 50% M Self-reported
Last 6 months Toska 2015 [54] 13.8% NR 2+ STI symptoms
Last 12 months Kembo 2012 [72] 5.4% NR Self-reported occurrence
Santelli 2013 [32] 25.5% 20.9% F; 37.5% M STI symptoms: genital ulcers
Santelli 2013 [32] 35.8% 42.6% F; 17.9% M STI symptoms: genital discharge
Santelli 2013 [32] 39.2% All female STI symptoms: vaginal discharge
Santelli 2013 [32] 64.2% All female STI symptoms: vaginal itching
Santelli 2013 [32] 16.2% All female STI symptoms: unpleasant vaginal odour
Santelli 2013 [32] 14.7% 16.9% F; 8.9% M STI symptoms: frequent urination
Santelli 2013 [32] 23.0% 22.3% F; 25% M STI symptoms: painful urination
Santelli 2013 [32] 11.3% 12.8% F; 7.1% M STI symptoms: pain during intercourse
Santelli 2013 [32] 1.5% 2.0% F; 0% M STI symptoms: bleeding during intercourse
Santelli 2013 [32] 29.4% 35.8% F; 12.5% M STI symptoms: lower abdominal pain
Santelli 2013 [32] 6.9% 8.1% F; 3.6% M STI symptoms: genital warts
Current STI Banura 2008 [57] 87.8% NR HPV–any infection (single or multiple strain–biomarker)
Banura 2008 [57] 64.6% NR HPV–single strain (biomarker)
Banura 2008 [57] 23.2% NR HPV–multiple strains (biomarker)
Katusiime 2012 [56] 6.1% NR HBV–positive for HBsAg (biomarker)
Pascoe 2015 [73] 49.7% NR HSV-2 infection (biomarker)
(Continued)

(NR–not reported)
Pregnancy Ever Baryamutuma 2010
[59]
13.2% NR Ever pregnant/ impregnated someone
Obare 2010 [118] 41% All female Ever pregnant among sexually active females
Birungi 2011 [61] 52% All female Ever pregnant
Gavin 2006 [68] 15% All female Ever pregnant: 13.7% among 10–14 y.o.; 20.6% among 15–19 y.o.
Mbalinda 2015 [52] 49% 56.9% F; 33.3% M Ever been or made someone pregnant
Nhamo 2014 [78] 40% All female At baseline of RCT
Obare 2010 [118] 60% 68% F; 27% M Ever been or made someone pregnant
Current Beyeza-Kashesya
2011 [60]
5% NR Pregnant at the time of the study
Gray 1998 [69] 15% All female 20.6% in 10–15 year olds; 13.7% in 20–24 year olds
Multiple pregnancies Birungi 2011 [61] 24.1% All female
Unintended pregnancy
Birungi 2011 [61] 73.9% All female Among those reporting at least one pregnancy
Nhamo 2014 [78] 75% All female Baseline of RCT, among all pregnancies
Obare 2010 [118] 75% NR
Safe sex Not reported Cataldo 2012 [28] Malawi: 29%; 53%; Mozambique: 20%; 73%;
Zambia: 33%; 70%; Zimbabwe: 10%-37%
NR Among sexually active participants only (rate among 10–14 year olds; rate among
15–19 year olds)
Highly protected sex Lightfoot 2007 [41] 69.5% NR Abstinent or consistent condom use at RCT baseline
Composite sexual risk- taking
HIV transmission risk
Senyonyi 2012 [42] 2.11 (SD 2.75) intervention group; 1.83 (SD
NR Score of several behaviours: number of sexual encounters
(intercourse or behaviour
2.57) control group
penetrative sex) + number of sexual partners + unprotected penile penetrative vaginal sexual acts
The included data on prevalence and potential factors associated with sexual risk-taking were cross-sectional (k =
31), while intervention data was based on a pre- and post-test experimental study (k = 1) and three RCTs (k = 3).
Sampling. Sampling strategies were assessed at two levels: community/clinics and individual level (Table 2). Of
the 31 observational studies, seventeen recruited only from healthcare facilities, primarily through purposefully
selected facilities (k = 14) [50,53,56–67]. Most of these studies recruited only HIV-positive participants (k = 11).
Seven other studies recruited only from communities through random or stratified sampling [29,68–73]. Most of
these studies recruited HIV-positive participants as part of larger community-based samples. Five studies
recruited participants through combined facility/community sampling [31,54,55,74,75].
At the individual-level sampling, eleven studies recruited through total or random sampling at each study site
[29,31,32,50,54,68–71,73,76], eight recruited through purposeful sampling [47,52,55,56,58,61–63], and another
eight through convenience sampling [28,53,57,59,60,66,74,75]. The three intervention studies which reported
sampling data recruited through a combination of purposeful and convenience sampling at both the facil-
ity/community and individual levels [40–42].
Sample sizes. The included studies reported on n = 13,536 HIV-positive adolescents and youth (10–24 years old),
with sample sizes ranging between n = 65 and n = 1,703. Fourteen of thirty-five studies had a sample size smaller
than n<400 participants, which was chosen as the cut-off for a study powered to detect predictors and intervention
effects, based on a recent systematic review which assessed studies of predictors of internalised HIV-stigma [49].
Studies with sample sizes <400 scored lower in the Study Quality Checklist.
Response and retention rate. Most studies (k = 26) did not report response or retention rates or did not have
response rates for the HIV-positive sub-population, making it difficult to assess the extent of selection bias. Of
studies that reported response or retention rates (k = 9), the majority stated retention of !90%
[41,42,53,54,60,74,75,77], with only one reporting a retention rate of 89.6% [28]. Two of the three small-scale
intervention studies analysed only data from completers, who accounted for 67.3% [42] and 59.6% [40] of those
who enrolled in the studies.

Strength of associations between factors/ interventions and outcomes. In the cross-sectional data analyses, 13
studies conducted univariate analyses (such as Chi square tests, univariate logistic regressions, or Student’s t-tests)
and eleven reported on multivariate analyses (such as logistic regressions, multivariate log binomial regression,
random effects logit model estimations) of associations between potential factors and the outcomes, controlling for
potential confounders (S6 Table). All four experimental design studies reported within-group change for at least
one sexual behaviour over time, with three randomising participants to a control and an intervention group
[41,42,78].
Prevalence of sexual risk-taking (Table 3, S6 Table)

Prevalence of high-risk sexual practices varied widely by outcome and should be interpreted with caution alongside
data on study methodology, sampling strategy and size, response rates, and definitions of outcome measures.
Nonetheless, several trends were notable. First, most studies reported high prevalence of sexual risk-taking in
relatively young populations (<19 years old). The prevalence of inconsistent condom use varied considerably,
though most of the studies found that between one-third and half of participants reported unprotected sex at last
intercourse. These prevalence rates are comparable to those reported by studies in the general adolescent
population in South Africa [71], Uganda [32], and other sub-Saharan African countries [79,80]. Of the studies
reporting current contraception use, one-third of the sample were on a form of contraception at the time of the study,
but only one study documented extremely low prevalence of dual protection– 5% of combined contraception and
condom use–in Uganda [60]. These results suggest that a high proportion of sexual acts by HIV-positive
adolescents and youth in sub-Saharan Africa are unprotected.
Second, where studies reported sexual debut, nearly half of the adolescents had had sex. This is an important
finding since most of the included participants were HIV-positive adolescents under 19 years old, which suggests
early sexual debut for a large part of the HIV-positive adolescent population. Early sexual debut has been linked to
reduced protected sex in the general adolescent population in South Africa [81]. Third, between half and two-thirds
of adolescents reported having an older sexual partner during first sex, which is linked to reduced condom use in
the general adolescent population [68]. Fourth, one in five participants reported engaging in transactional sex or
having sex for money or goods–a sexual practice associated with unprotected sex [82], a result that was consistent
across multiple studies [63,73,75]. HIV-positive adolescent girls reported a higher prevalence of transactional sex,
having older sexual partners, and unprotected sex, though adolescent boys were more likely to report early sexual
activity and multiple sexual partners.
Fifth, the prevalence of multiple sexual partners varied widely, though most studies reported that at least one in
ten participants had had at least two sexual partners in the last 12 months and one in three male participants reported
multiple sexual partners [29,32,83]. Sixth, studies testing for biomarkers of HSV-2 and HPV found 50% and 88%
infection rates in HIV-positive adolescents, respectively [57,73]. This suggests high levels of unprotected sex and
higher risk for co-infection with other sexually transmitted infections.
Factors associated with sexual risk-taking among HIV-positive adolescents and youth
(Table 4)
Seventeen of the included studies (23 publications) reported associations between one or more factors and sexual
outcomes, with univariate and multivariate analyses testing the strength of these relationships. No longitudinal
predictors or causal relationships between factors and sexual risk-taking were reported by any of the included
studies. Potential factors associated with

Table 4. Factors associated with sexual-risk taking reported in included studies.
Factor level/ grouping Factor Increased sexual risk-taking Non-significant associations Decreased sexual
risk-taking
Individual–Socio- demographic factors
Age (older) Ever had sex [52] Condom use [64], unprotected sex [54], multiple
sexual partners [64,76]
Unprotected sex [76], condom use [64,77], contraception use1 [51]
Gender (female) Unprotected sex [54,76], contraception use [51],
multiple sexual partners[76]
Rural residence Unprotected sex [54]
Informal housing Unprotected sex [54]
Study site Contraception use [61], unintended pregnancy [61],
unprotected sex [76], multiple sexual partners [76]
Individual–Mental and physical health factors
Depression (clinical) Multiple sexual partners [64] Condom use [64]
Anxiety Condom use [64], multiple sexual partners [64]
Poor birth outcomes Contraception use [61]
Individual–Knowledge, attitude, and beliefs
Does not drink alcohol Multiple sexual partners [76] Ever had sex [52], unprotected sex
[76]
STI prevention knowledge Unprotected sex [76], multiple sexual partners [76] Relationship-level
factors Has children with husband Unintended pregnancy [61] Contraception use3 [61]
Living arrangement: lives with partner Contraception use [51],
unintended pregnancy [87]
Gender-based violence Unintended pregnancy [61,87],
multiple sexual partners [87]
Family and community- level factors
Lives with biological parent Unprotected sex [54]
Lives alone Ever had sex [52]
Orphanhood Unprotected sex [54,63] Parental monitoring4social support Unprotected sex2 [63]
Social support Condom use [64], multiple sexual partners [64]
Structural-level factors Education Unprotected sex [76], multiple sexual partners [76] Ever had sex [114]
Maternal education Contraception use3 [61], unintended pregnancy
[61]
Poverty Unprotected sex [54]
Food insecurity Unintended pregnancy [87]
Employment Ever had sex [52]
Intervention–combination social protection (grants + livelihood training + SRH services)
Multiple sexual partners [87] Condom use [87], transactional sex
[87]
HIV-related factors Knows own HIV+ status Unprotected sex [54]
Mode of infection (vertical) Condom use [64], multiple sexual partners [64] Unprotected sex [54]
Time since diagnosis (years)
Time on ART Unprotected sex [54]
ART adherence Unprotected sex [54]
Opportunistic infections Unprotected sex [54]
Partner HIV-status unknown Multiple sexual partners [76] Unprotected sex [54,76]
Disclosed HIV status to partner Unprotected sex [54]
ART use/ access Unprotected sex [76], condom use [64], multiple
sexual partners [64,76]
ART care (hospital vs. primary clinic) Unprotected sex [54]
Intervention–access to health services: HIV support group
Condom use [40]
1 All contraception-related outcomes are included: at first sex, ever used any modern method, and current use of any
method. 2 Study reported on the individual factors in univariate analyses, neither of which were significant. The
interaction term was also not significant at p = 0.11. 3 Post-partum contraception.
sexual risk-taking among HIV-positive adolescents and youth are presented in five groups following the
socio-ecological model [84–86] as a theoretical framework (Fig 2): (1) individual- level factors, (2)
relationship-related factors, (3) family and community factors, (4) structural factors, and (5) HIV-specific factors.
Nine studies included associations between risky sexual practices [52,54,59,61,63,75–77,87]. Most associations
between different types of sexual risk- practices were not statistically significant, with. only two studies reporting
significant multivariate associations using cross-sectional data: inconsistent condom use and having multiple

Fig 2. Hypothesised factors associated with sexual risk-taking among HIV-positive adolescents and youth in
Sub- Saharan Africa.
https://doi.org/10.1371/journal.pone.0178106.g002

sexual partners were associated with unintended pregnancy [61,87]. Only multivariate results are described in detail
in this section with all associations and actual statistical test results– where available–included in S6 Table.
Individual-level factors. Seventeen studies assessed individual-level factors using multivariate analyses.
Socio-demographic factors included: age [50,52,54,64,74,76], gender [51,54,76], rural residence [54], and informal
housing [54]. Mental and physical health factors included: clinical depression [64], anxiety [64], and having poor
birth outcomes, such as pre- mature birth, small birth weight, and child being small for gestational age [61].
Knowledge, attitude, and behaviours included drinking alcohol [50,76], and STI prevention knowledge [76]. Four
studies reported inconsistent associations between age and sexual risk-taking: three reported that older age is
associated with lower reports of high-risk sexual practices [64,74,76], but no significant associations with others
[54,64,76]. Three studies reported non-significant multivariate analyses on the relationship between gender and
sexual risk-taking [54,74,76], although gender-disaggregated prevalence rates suggest that HIV-positive adolescent
girls/ young women engage in higher levels of risk-taking compared to HIV-positive adolescent boys/ young men
[28,68,69]. Rural residence and informal housing were not significantly associated with any sexual practices.
Two studies reported mental and physical health factors associated with sexual risk-taking: depression, anxiety,
and poor birth outcomes. Clinical depression was significant associated with lower condom use but not with having
multiple sexual partners [64]. Anxiety and poor birth outcomes were not significantly associated with any of the
outcomes [61,64].
Two studies tested multivariate associations between knowledge, attitudes, and behaviours potentially linked with
sexual risk-taking. Of these, two found that adolescents who reported alcohol-drinking were more likely to be
sexually active and less likely to use condoms [52,76], but no statistically significant association between drinking
alcohol and reporting multiple sexual partners [76]. STI prevention knowledge were not significantly associated
with unprotected sex [76].
Relationship factors. Relationship-level factors tested by three studies included having a child with one’s
husband, living with a partner (compared to at home with parents/ caregivers), and gender-based violence
[51,61,87]. HIV-positive adolescents and youth were more likely to report unintended pregnancy if they had
biological children with their husband [61]. They were more likely to report multiple sexual partners and unintended
pregnancies if they lived with their partner [87] or experienced gender-based violence [61,87].
Family and community factors. Four studies tested associations between four family and community factors
associated with one or more sexual risk practices [52,54,63,64] using multivariate analyses, including: living with
at least one biological parent, orphanhood, and parenting relationship (monitoring), and having a supportive
family. Ugandan adolescents living alone were more likely to be sexually active in one study [52]. All other factors
were not significantly associated with sexual risk-taking.
Structural factors. Six studies (five cross-sectional and one intervention) tested the effects of six structural factors
or provisions on sexual risk-taking using multivariate analyses: poverty, food insecurity, participant employment,
education (adolescent and mother), accessing grants and receiving livelihood training.
Employed adolescents were more likely to have ever had sex even when taking age into account [52]. Food
insecurity was strongly associated with unintended pregnancies in the baseline sample of an RCT for HIV-positive
orphaned and out-of-school adolescent girls in Zimbabwe [87]. In the same study, accessing grants in combination
with health services and vocational training was associated with increased condom use and reduced transactional
sex, but no reduction in multiple sexual partners was documented [78]. Findings on access to

education were inconsistent with maternal education and poverty not associated with sexual risk-taking.
HIV-related factors. Four studies tested associations between fourteen HIV-related factors and sexual
risk-taking practices using multivariate analyses. Knowing one’s own HIV-positive status and access to HIV
support groups were associated with reduced unprotected sex. Findings on mode of infection and knowledge of
partner’s HIV status were inconsistent [54,64,76]. Time since HIV diagnosis, time on ART, ART adherence,
reporting opportunistic infections, disclosing HIV-status to partners, ART use/ access, and receiving ART care at a
hospital were not significantly associated with unprotected sex [54,64].
Interventions addressing sexual risk-taking among HIV-positive adolescents and youth

This review located eight studies of interventions aiming to address one or more sexual risk practices among
HIV-positive adolescents or youth [88–93], but data on intervention effects were only available for four
interventions, which are included in this review [40–42,78]. Three of the included intervention studies were
individual-level randomised controlled trials [41,42,78] (Table 5), with one pre- and post-test experimental design
study [40]. All interventions measured at least one high-risk sexual practice, with all measuring condom use or
unprotected sex, three measuring number of sexual partners, and one measuring transactional sex. One of the
studies reported effects on a composite sexual behaviour transmission score [42]. All four studies reported increases
in condom use following the intervention; however, in one study the increase was not significant when intervention
and control groups were compared post-intervention [42].
Among the three interventions that had significant effects on reducing sexual risk-taking, one was an
individual-based 18-session intervention delivered by nurses (n = 50 intervention and n = 50 control) [41], the
second a group-based intervention focused on improving self- efficacy (n = 65) [40], and a third focused on
addressing both individual and structural drivers of risk-taking through group-based life-skills training and
livelihood support through grants and vocational training (n = 710) [78]. Though the first two studies are promising,
the studies testing them reported on small sample sizes (n<120). To test their generalisability and scalability, these
interventions need to be replicated with larger samples in real-life settings. The only large-scale intervention
included in the review reported positive results from an intervention combining HIV/STI health services, life skills
training, and livelihood components, such as small grants [78]. Findings from the combination intervention
suggested that the results were due to the combination intervention rather than the life skills component alone [78].
This intervention focused on highly vulnerable HIV-positive adolescent girls, so additional research on whether
these interventions may work among HIV-positive adolescents and youth in general would be beneficial.
Discussion
This review includes 35 studies documenting the prevalence of sexual risk-taking, factors associated with high-risk
sexual practices, and interventions for reducing sexual risk-taking in HIV-positive adolescents and youth from 13
sub-Saharan African countries. All studies reported on prevalence of high-risk sexual practices, and sixteen reported
on at least one potential factor associated with sexual risk-taking. Four studies reported on interventions to reduce
sexual risk-taking among HIV-positive adolescents. This section summarises the implications of the quality of
included studies, followed by recommendations for a research agenda on the sexual practices of HIV-positive
adolescents and youth.

Table 5. Summary of included intervention studies.
Author Year, Country
Study design Follow-up time Sample1 SQC
Outcomes Results score2
Lightfoot 2007 [41], Uganda
Intervention name, content, delivery mode
Number of sexual partners: Log number of sexual partners decreased (F 1,19 = 4.68, p = 0.04) Consistent condom
use: Consistent condom use increased from 10% to 93% in intervention (p<0.01), control did not significantly change,
from 15 to 12%. Highly protected (abstinent or consistent condom use): in intervention arm significantly increased
from 74% to 98% (p<0.01), no change in control from 65% to 62%, NS.
Nhamo 2014 [78], Zimbabwe
Cognitive Based Therapy
RCT
Assessed at baseline
14–21 year
75% Number of sexual partners (CBT)
Intervention: 50
and
olds
Consistent condom use One-on-one with
nurses
Control: 50
3 months
72% female
Highly protected sex (abstinence 18
sessions
Retention rate:
Not reported
or consistent condom use) 90%
Multiple sexual partners: changed from 6% to 7%, OR = 1.05, .90–1.23, p = 0.504 Transactional sex: changed from
60% to 49%, OR = 0.87, 0.75–1.01, p = 0.067 Condom use: RR = 1.43 95% CI1.16–1.76, p<0.001
Senyonyi 2012 [42], Uganda
Shaping the Health of
RCT pre- and
Baseline and follow-up
16–19 years
66% Multiple sexual partners Adolescents in
Zimbabwe
post-test
at 6, 12, and 18
old
Transactional sex (SHAZ)-Plus!
N = 710
months; sexual
100% female
Condom use 0–6 months: both arms
Retention rate not
outcomes reported for
Mode of receive HIV/ SRH services
reported
pre- test (baseline) to
infection not + Life Skills education
post-test (18 months)
reported; 7–12 months: intervention
orphaned and receives HIV/ SRH services + livelihood
intervention (vocational training &
out of school, not pregnant at enrolment grant); control receives only HIV/SRH services 13–18 months: both groups
receive only HIV/ SRH services including testing
CBT
RCT
No information 12–18 years
55% Sexual transmission
Sexual
transmission behaviour Group counselling delivered
328 contacted;
53% female
score = number of sexual
score: decrease
in total score for by trained counsellors
171 selected to
Vertically
encounters (intercourse or
both the
intervention and control Recurring weekly
participate in
acquired HIV
penetrative sex), number of sexual
groups, Wilk’s λ = 0.951, F
1
,113 sessions
intervention
partners, and unprotected penile 80 min
per session
N = 115
penetrative vaginal sexual acts completed 3
(intercourse; i.e., use of condoms, + sessions: n = 80
and continued abstinence) intervention, n = 35 control
= 5.866, p = 0.017, partial η2 = 0.049. Repeated measures ANOVA showed no significant
group differences at post-test, Wilk’s λ = 1.00, F(1,113) = 0.024, p = 0.876, partial η2 0.001.
Snyder 2014 [40], South Africa
Hlanangani
Pilot study
Start of session 1
16–24 years
41% Condom use in previous 3 weeks
Condom use:
increased by 12% CBT
n = 109, 74 (68%)
(baseline) to
95% female
(during intervention)
(p = 0.049).
Social Cognitive Theory
returning for all
end of session 3
Unknown Support groups delivered
three sessions,
(follow-up).
mode of by lay counsellors
analyses of
Analyses n = 65
infection 3 (2 hour) sessions over 11
completers
Past-year months
diagnosis
1 When available, age, gender, mode of infection, and other inclusion criteria are included. 2 Study Quality Checklist
score.
Quality of included studies

Most studies scored <50% in the Study Quality Checklist due to low scores on whether studies could establish
causality. Notably, this was due to more than half the studies recruiting HIV- positive adolescents and youth as part
of larger community-based samples, and not conducting analyses on the HIV-positive sub-sample, as this was not
part of the original study aims. These datasets represent a unique opportunity to conduct secondary data analyses of
the HIV-positive sub-samples in these studies to provide additional–more generalizable–insights from com-
munity-based samples. Most of the included studies focused on HIV-positive adolescents and youth recruited from
facility-based samples. By engaging mostly adolescents who are already receiving care and support, the included
studies are likely to under-estimate the prevalence of

sexual risk-taking, as adolescents who are not engaged in care are more likely to experience poor sexual health
outcomes [94]. A further limitation of the quality of included studies is that the majority of studies reported on
cross-sectional analyses, which limited our ability to reach any conclusions about the causality and strength of the
reported relationships between hypothesised factors and high-risk sexual practices. Rigorous longitudinal research is
needed to better understand and test pathways of increased sexual risk-taking among HIV-positive adolescents and
youth in sub-Saharan Africa.
Further, nearly half of the studies that tested hypothesised factors reported only on univariate analyses, mostly
Chi square tests. A systematic review of condom use studies among adolescents in the United States also found
that the majority of reviewed studies utilised univariate rather than multivariate analyses [95]. Theory- and
policy-driven multivariate analyses are needed to gain a clearer picture of factors associated with sexual risk taking
in HIV-positive adolescents and youth and how they are influenced by confounding factors. Quasi-experimental
analyses can then be applied to assess which modifiable factors can be changed to reduce sexual risk-taking in this
highly vulnerable population [96,97]. The poor quality of the studies and limited analytic techniques utilised may
also explain why most of the reported findings were inconsistent.
Implications for future research and practice

None of the included studies assessed longitudinal predictors, making it difficult to inform the design of large-scale
policies and programming for HIV-positive adolescents and youth. How- ever, the findings of this review have
several implications for policy and practice.
First, HIV-positive adolescents and youth are reporting high levels of sexual risk-taking, which could lead to
passing the virus to uninfected partners and children. These rates are high even though most of the included
HIV-positive adolescents and youth were already in treatment and care. This suggests that improved sexual and
reproductive health services for HIV- positive adolescents should be provided as part of regular HIV treatment and
care. Successful models of integrating sexual and reproductive health services with HIV treatment and care should
be designed, tested, and rolled out. Some adolescent- and youth-friendly models applied in resource-poor settings
from organizations such as the Paediatric AIDS Treatment for Africa and its collaborators [98] should be further
explored.
Second, nearly three-quarters of all pregnancies among HIV-positive adolescent women were reported to be
unintended [53,74,78], higher than reported rates of unintended pregnancies among adolescent girls in South
Africa [99]. While unintended pregnancy is not a risk practice itself, being pregnant during adolescence for
HIV-positive adolescents presents a two- fold risk. On the one hand, adolescent pregnancy and motherhood has been
linked to poorer health, educational, and socio-economic outcomes for both mother and child [99–102]. On the other
hand, children of HIV-positive adolescent mothers are at increased risk of mother- to-child-transmission, due to poor
adherence to ART and retention in care [14]. Reducing unintended pregnancies can have great positive effects in
improving maternal and child health outcomes, particularly among HIV-positive adolescents [103]. The prevalence
of unintended pregnancies among HIV-positive adolescents and youth suggest that the unmet contraceptive need
among this population is even higher. Potential barriers to accessing contraception include quality of sexual and
reproductive health services received at health facilities [25], stigma and discrimination [26,94], and lack of
integration of family planning and HIV care services [104]. Additional research on how HIV-positive adolescent
girls/ youth experience antenatal care services, PMTCT, and early parenthood is urgently needed as this cohort
comes of age and enters their reproductive years.

Third, HIV-positive adolescents who are most vulnerable may also be at highest risk for onwards HIV
transmission. Although most associations reported in the included studies were inconsistent, this review’s findings
suggest that policies and programming must take into con- sideration the unique needs and profile of HIV-positive
adolescents. In nine studies, sexual health outcomes were associated with other sexual risk-taking practices. For
example, unintended pregnancy was associated with having multiple sexual partners. Only one intervention aimed
to reduce multiple risk-taking, with no significant results between intervention and control groups [42]. Better
understanding of factors shaping the syndemic of risk-taking among HIV-positive adolescents and youth will be
important to help healthcare providers screen for and support those adolescents most at risk. Practitioners should be
sensitised and trained to deal with the specific needs of HIV-positive adolescents and youth, as they initiate sexual
and romantic relationships, and enter their reproductive years.
Fourth, the evidence on potential interventions, although based primarily on small-scale trials, suggests that
improved non-medical care, psychosocial care and support, may reduce sexual risk-taking among HIV-positive
adolescents. Testing the scalability of low-cost interventions delivered by lay counsellors such as support groups
will be key to evidence-based programming for HIV-positive adolescents. HIV-specific support groups have had
encouraging results in improving HIV treatment outcomes among adolescents and youth in other parts of South
Africa [40,105], in neighbouring countries [106,107], as well as in this sample [108]. Finally, the results of the
largest randomised controlled trial included in this review (n = 710) suggest that combinations of interventions may
have a stronger impact on reducing sexual risk-taking among HIV-positive adolescents and youth [78]. Similarly to
primary HIV prevention efforts, potential combinations of behavioural, biomedical and structural interventions
should be tested [109,110]. Complex multi-level interventions require comprehensive financial, capital, and human
resources to be designed and conducted. Therefore, testing combinations of existing low-resource real-life
interventions, potentially using pragmatic research methodologies such as quasi-experimental designs, should
precede setting up and testing potential combinations of interventions through randomised controlled trials.
Based on the quality of the evidence base, additional foundational and intervention research is needed to inform
interventions for HIV positive teens. Evidence-based policy and programming could benefit from additional
research with representative samples of HIV-positive adolescents and youth. The following research gaps were
identified through this systematic review:
High-risk sexual practices and exposure to violence: This review has highlighted several gaps in research. First,
better understanding of adolescents who are at risk of engaging in multiple high-risk sexual practices is pivotal to
improved care and support programming. Second, although this systematic review did not focus on adolescent
experiences of violence such as forced sex, gender-based violence, and domestic violence, multiple of the included
studies found high rates of forced sexual initiation and gender-based violence amongst HIV-positive adolescents
[74,75,87]. In light of the growing evidence linking exposure to violence and abuse among adolescents with poor
health outcomes [111,112], it is important to understand whether and how violence exposure is associated with
high-risk sexual practices in HIV-positive adolescents.
Under-studied potential factors shaping sexual risk-taking: A recently updated systematic review reports that
vertically infected HIV-positive adolescents experience significant neurocognitive delays [18]. These findings are
confirmed from research in South Africa, which has documented significant brain damage due to vertical HIV
infection [113]. However, none of the included studies recorded whether the HIV-positive participants had
neurocognitive problems, which may shape how they engage in sexual relationships. Only one study reported

on mental health outcomes, controlling for sexual health outcomes [114], with data from a second study testing
whether depression or anxiety was associated with sexual risk-taking. Fur- ther research is needed to elucidate how
HIV-specific mental health issues affect sexual practices among HIV-positive adolescents and youth.
Mode of infection may be an important determinant of risk taking, with horizontally infected adolescents
continuing to engage in higher risk-taking following HIV infection. Most of the studies in this review that reported
mode of infection recruited primarily vertically infected adolescents. Though mode of infection is often difficult to
ascertain and record [115], future studies should attempt to document factors that could ascertain the participants’
mode of infection, such as time on treatment, age at HIV diagnosis, and maternal orphanhood [10]. Additional
research is needed on longitudinal predictors and interventions for horizontally infected adolescents and youth. Few
studies reported age-disaggregated data, which would help providers to better understand how to tailor services and
programmes to different age groups and to adapt to the changing needs of growing HIV-positive adolescent
populations.
Studies of older adolescents and youth recorded several key relationship factors, such as whether the
HIV-positive participant lived with their partners, had children with their husband vs. boyfriend, wanted additional
children, or had disclosed their status in the relationship. However, most of the findings on these relationship topics
were inconsistent. As HIV- positive adolescents get older and navigate sexual and romantic relationships, additional
research is needed to understand how HIV-positive status, its disclosure and other relationship-related factors
interact to shape sexual and reproductive outcomes [43,116,117].
Four studies [31,60,67,118] recruited a comparison sample of HIV-negative or HIV-status- unknown adolescents
or youth, with three documenting how HIV-status knowledge was associated with sexual health outcomes
[31,60,118]. Though this comparison was beyond the scope of this review, HIV-positive status was associated with
increased protective sexual practices in one study and increased risk-taking in two other studies. These findings
confirm those of a similar review which included studies from North America and Europe [30] but did not report
consistent associations between knowledge of HIV-positive status and sexual practices. Find- ings from a nationally
representative community-based study in South Africa suggest that knowledge of HIV-status, whether positive or
negative, is strongly associated with reduced sexual risk-taking [29]. Further analyses on the effect of knowing
one’s HIV-positive status on sexual risk-taking and the mechanisms through which disclosure shapes sexual
practices is needed, particularly in light of the UNAIDS Fast Track target of 90% of HIV-positive adolescents
learning their status by 2020 [7].
Intervention research gaps: First, although results from three cognitive-based therapy trials were encouraging,
they need to be tested in larger-scale trials. Second, the majority of the studies in this review (17 of 35) took place in
Uganda, including three of the four interventions. Additional quantitative and intervention research is needed on
secondary HIV prevention in other sub-Saharan countries, particularly South Africa, Nigeria, and Kenya—home to
the largest populations of HIV-positive adolescents in the region [5]. Third, several family-level factors were
associated with reduced risk-taking in cross-sectional univariate analyses [52,59,63,74]. Although adolescence is a
time for exploring and testing boundaries [119], family-based interventions may be key to support with difficult
HIV-related issues such as disclosure, transition in care, and accessing prevention-of-mother-to-child treatment
[120,121]. Fourth, in-depth analyses are needed of linkages between healthcare experiences (quality of services,
access to services such as family planning, counselling and support groups) of HIV-positive adolescents and youth
and their sexual and reproductive practices. Evidence to date is limited and inconsistent on the role of group-based
care and support, though promising evidence from several programmes has been documented in recent conferences
and workshops [105–107,122].

Review limitations
In addition to the research gaps identified above, this review had several limitations. First, it included multiple
outcomes to measure sexual risk-taking. Although evidence on linkages between sexually transmitted infections and
high risk practices in adolescents are established [32,123,124], evidence that this review’s outcomes are associated
with secondary HIV transmission is limited. Second, studies varied widely in terms of sample size, sampling
strategies, and exact definitions of outcome measures, and the majority of studies were cross-sectional. Therefore, a
meta-analysis was not possible, and our ability to reach conclusions on the prevalence and factors of sexual
risk-taking among HIV-positive adolescents and youth was limited. Third, the analyses reported by the included
studies were mostly univariate with actual statis- tics often not reported and confidence intervals missing, which
resulted in a low quality of included evidence. Finally, the majority of studies were conducted in Uganda, including
mostly HIV-positive adolescents and youth in care. Therefore, the results are not generalizable across the whole
HIV-positive adolescent and youth population in sub-Saharan Africa. The evidence presented here must be
interpreted with these methodological limitations in mind.
Conclusion
HIV-positive adolescents have been neglected in HIV prevention efforts in the region, with few studies testing
interventions aimed at supporting HIV-positive adolescents to reduce sexual and onwards vertical transmission
(secondary prevention). Very few studies have rigor- ously documented potential risk and protective factors
associated with increased secondary HIV-transmission risk. Longitudinal research is needed to establish and test
emerging patterns between HIV-transmission risk and socio-demographic, HIV-specific, relationship, family, and
structural-level factors. To address the potential for onwards HIV transmission, evidence is urgently needed on the
effectiveness and feasibility of low-cost interventions to reduce HIV transmission from adolescents, both vertically
and horizontally infected.
Supporting information
S1 PRISMA Checklist. PRISMA checklist. (DOC)
S1 File. Data extraction form. (DOCX)
S2 File. Study quality checklist and risk assessment bias. (DOCX)
S1 Table. Study inclusion and exclusion criteria. (DOCX)
S2 Table. Search string for databases searched in OvidSP. (DOCX)
S3 Table. Search strings for PubMed & Proquest. (DOCX)
S4 Table. Search strings for other smaller databases. (DOCX)
S5 Table. Study screening checklist. (DOCX)

S6 Table. Results of factors associated with sexual risk-taking among HIV-positive adolescents and youth by
study. (DOCX)
S7 Table. Prevalence of sexual risk-exposure outcomes reported by included studies. (DOCX)
S8 Table. Results of the risk of bias assessments for all included studies. (DOCX)
Acknowledgments
The authors would like to thank the authors of the primary research studies from sub-Saharan Africa for their
dedication to the sexual health and well-being of HIV-positive adolescents and youth in the region. We especially
thank the researchers who sent additional data and information to complete this review: Prof. Renee Heffron, Dr.
Irving Hoffman, Dr. Kimberly Powers, Dr. Edna Viegas, Dr. Christiana No ̈stlinger, Dr. Olive Shisana, Dr.
Kangelani Zuma, and Goit- seono Mafoko, Dr. Karine Dube, Dr. Vicky Jespers, Prof. Anna-Lise Williamson, Dr.
Leigh Johnson, Dr. Gabriela Paz-Bailey, Nicola Willis, Dr. Webster Mahvu, Edward Pettitt II.
Author Contributions
Conceptualization: ET MP FM LC.
Data curation: ET MP FM RH KK.
Formal analysis: ET.
Funding acquisition: ET LC.
Investigation: ET KK MP FM RH.
Methodology: ET MP FM LC.
Project administration: ET.
Validation: MP FM KK RH.
Visualization: ET.
Writing – original draft: ET.
Writing – review & editing: ET MP FM RH.
References
1. Ferrand RA, Corbett EL, Wood R, Hargrove J, Ndhlovu CE, Cowan FM, et al. AIDS among older chil-
dren and adolescents in Southern Africa: projecting the time course and magnitude of the epidemic. AIDS. 2009; 23:
2039–2046. https://doi.org/10.1097/QAD.0b013e32833016ce PMID: 19684508 2. UNAIDS. Fact sheet 2015 |
UNAIDS [Internet]. Geneva, Switzerland; 2015. http://www.unaids.org/en/
resources/campaigns/HowAIDSchangedeverything/factsheet 3. Africa’s Health Project. Improving Adolescent
HIV Treatment, Care, Prevention and Family Planning Services [Internet]. 2012.
http://www.aidstar-one.com/sites/default/files/ALHIVtechbrief2012July10_ AFRbureau.pdf 4. UNAIDS. The gap report
[Internet]. Geneva, Switzerland; 2014. http://www.unaids.org/sites/default/
files/media_asset/UNAIDS_Gap_report_en.pdf

5. Kassede S. Adolescents living with HIV: A global perspective and lessons from country experience in
sharpening national response [Internet]. New York, USA: UNICEF; 2015. https://aidsfree.usaid.gov/
sites/default/files/1_susan_kassede_alhiv.pdf 6. UNAIDS, AU. Empower young women and adolescents girls:
Fast-tracking the end of the AIDS epi-
demic in Africa [Internet]. Geneva, Switzerland; 2015. http://www.unaids.org/sites/default/files/media_
asset/JC2746_en.pdf 7. UNAIDS. 90-90-90 An ambitious treatment target to help end the AIDS epidemic [Internet].
Geneva,
Switzerland; 2014. p. 40. http://www.unaids.org/sites/default/files/media_asset/90-90-90_en_0.pdf 8. Ajose O,
Mookerjee S, Mills EJ, Boulle A, Ford N. Treatment outcomes of patients on second-line antiretroviral therapy in
resource-limited settings: a systematic review and meta-analysis. AIDS. 2012; 26: 929–38.
https://doi.org/10.1097/QAD.0b013e328351f5b2 PMID: 22313953 9. WHO. HIV drug resistance report 2012
[Internet]. 2012. Report No.: QV 268.5. http://www.who.int/hiv/
pub/drugresistance/report2012/en/ 10. Sharer M, Fullem A. Transitioning of Care and Other Services for
Adolescents living with HIV in Sub- Saharan Africa. USAID’s AIDS Support and Technical Assistance Resources,
AIDSTAR-ONE, Task Order 1. Arlington, VA; 2012. 11. Hudelson C, Cluver LD. Factors associated with adherence to
antiretroviral therapy among adoles-
cents living with HIV/AIDS in low- and middle-income countries: a systematic review. AIDS Care. 2015; 27: 805–816.
https://doi.org/10.1080/09540121.2015.1011073 PMID: 25702789 12. Nachega JB, Hislop M, Nguyen H, Dowdy DW,
Chaisson RE, Regensberg L, et al. Antiretroviral Ther-
apy Adherence, Virologic and Immunologic Outcomes in Adolescents Compared With Adults in South- ern Africa. J
Acquir Immune Defic Syndr. 2009; 51: 65–71. https://doi.org/10.1097/QAI. 0b013e318199072e PMID: 19282780 13.
Evans D, Menezes C, Mahomed K, Macdonald P, Untiedt S, Levin L, et al. Treatment Outcomes of
HIV-Infected Adolescents attending Public-Sector HIV Clinics across Gauteng and Mpumalanga, South Africa. AIDS
Res Hum Retroviruses. 2013; 29: 892–900. https://doi.org/10.1089/AID.2012. 0215 PMID: 23373540 14. Fatti G,
Shaikh N, Eley B, Jackson DJ, Grimwood A. Adolescent and young pregnant women at
increased risk of mother-to-child transmission of HIV and poorer maternal and infant health outcomes: A cohort study
at public facilities in South Africa. South African Medical Journal. 2014. pp. 874–880. PMID: 26042271
15. Chawana TD, Ngara B, Katzenstein D, Nathoo K, Nhachi CFB, (ATF) ATF study. An adherence inter-
vention and drug resistance among HIV positive adolescents failing 2nd-line treatment at a public health clinic-a
randomised controlled trial. 21st International AIDS Conference—AIDS2016. Durban, South Africa: International
AIDS Society; 2016.
16. Schramm B, Carnimeo V, Rakesh A, Ardiet DL, Cossa L, Bouchaud O, et al. Cross-sectional assess-
ment of virological failure, drug resistance and third-line regimen requirements among patients receiving second-line
ART in 3 large HIV-programmes in Kenya, Malawi and Mozambique. 21st International AIDS Conference (AIDS
2016). Durban, South Africa: International AIDS Society; 2016. 17. Lowenthal ED, Bakeera-Kitaka S, Marukutira T,
Chapman J, Goldrath K, Ferrand RA. Perinatally
acquired HIV infection in adolescents from sub-Saharan Africa: A review of emerging challenges. The Lancet
Infectious Diseases. 2014. pp. 627–639. https://doi.org/10.1016/S1473-3099(13)70363-3 PMID: 24406145 18. Sherr
L, Croome N, Parra Castaneda K, Bradshaw K, Herrero Romero R. Developmental challenges
in HIV infected children—An updated systematic review. Child Youth Serv Rev. 2014; 45: 74–89.
https://doi.org/10.1016/j.childyouth.2014.03.040
19. Amzel A, Toska E, Lovich R, Widyono M, Patel T, Foti C, et al. Promoting a combination approach to
paediatric HIV psychosocial support. AIDS. 2013; 27 Suppl 2: S147–57. https://doi.org/10.1097/QAD.
0000000000000098 PMID: 24361624 20. Wiener LS, Battles HB. Untangling the web: a close look at diagnosis
disclosure among HIV-infected
adolescents. J Adolesc Heal. 2006; 38: 307–309.
21. Mellins CA, Bhana A, Petersen I, Holst H, Alicea S, Myeza N, et al. The VUKA family project: a family- based
mental health and HIV prevention program for perinatally HIV-positive youth. 19th International AIDS
Conference—AIDS2012. Washington D.C., USA; 2012. 22. Busza J, Besana GVR, Mapunda P, Oliveras E. “I have
grown up controlling myself a lot.” Fear and
misconceptions about sex among adolescents vertically-infected with HIV in Tanzania. Reprod Health Matters. 2013;
21: 87–96. https://doi.org/10.1016/S0968-8080(13)41689-0 PMID: 23684191 23. Shisana O, Rehle TM, Simbayi LC,
Zuma K, Jooste S, Pillay-van-Wyk V, et al. South African National
HIV Prevalence, Incidence, Behaviour and Communication Survey 2008—A Turning Tide Among Teenagers? Cape
Town: HSRC Press; 2009.

24. Jaspan HB, Li R, Johnson L, Bekker L-G. The Emerging Need for Adolescent-Focused HIV Care in
South Africa. South Afr J HIV Med. 2009; 9–11. 25. Wood K, Jewkes RK. Blood Blockages and Scolding
Nurses: Barriers to Adolescent Contraceptive
Use in South Africa. Reprod Health Matters. 2006; 14: 109–118. https://doi.org/10.1016/S0968-8080 (06)27231-8
PMID: 16713885 26. Mburu G, Hodgson I, Teltschik A, Ram M, Haamujompa C, Bajpai D, et al. Rights-based
services for
adolescents living with HIV: adolescent self-efficacy and implications for health systems in Zambia. Reprod Health
Matters. 2013; 21: 176–185. https://doi.org/10.1016/S0968-8080(13)41701-9 PMID: 23684200 27. Busza J. Meeting
the Reproductive Health Needs of Youth Living with HIV in Tanzania: A qualitative
study exploring the experiences and perceptions of young home based care clients, their caregivers, and care
providers. 2011. 28. Cataldo F, Malunga A, Rusakaniko S, Umar E, Zulu J, Teles N, et al. Experiences and
Challenges in
Sexual and Reproductive Health for Adolescents Living with HIV in Malawi, Mozambique, Zambia and Zimbabwe.
19th International AIDS Conference—AIDS2012. Washington D.C., USA; 2012. 29. Shisana O, Rehle TM, Simbayi
LC, Zuma K, Jooste S, Zungu NP, et al. South African National HIV
Prevalence, Incidence and Behaviour Survey, 2012. HSRC Press. Cape Town, South Africa; 2014. 30. Mergui
A, Giami A. La sexualite ́ des adolescents se ́ropositifs: analyse de la litte ́rature et re ́flexion sur
les impense ́s de la sexualite ́ [The sexuality of HIV-infected adolescents: Literature review and thinking the
unthinkable of sexuality]. Arch Pediatr. 2011; 18: 797–805. https://doi.org/10.1016/j.arcped.2011. 04.015 PMID:
21652188 31. Morris L, Kouya F, Kwalar R, Pilapil M, Saito K, Palmer N, et al. A comparison of high risk behaviours
in HIV-positive, HIV-negative and HIV status unknown youth in Cameroon. 19th International AIDS
Conference—AIDS2012. 2012. 32. Santelli J, Edelstein ZR, Mathur S, Wei Y, Zhang W, Orr MG, et al. Behavioral,
Biological, and Demo- graphic Risk and Protective Factors for New HIV Infections Among Youth in Rakai, Uganda. J
Acquir Immune Defic Syndr. 2013; 63: 393–400. https://doi.org/10.1097/QAI.0b013e3182926795 PMID: 23535293
33. Butler RB, Schultz JR, Forsberg a D, Brown LK, Parsons JT, King G, et al. Promoting safer sex among
HIV-positive youth with haemophilia: theory, intervention, and outcome. Haemophilia. 2003; 9: 214– 222. PMID:
12614374
34. Chen X, Murphy DA, Naar-king S, Parsons JT, and the Adolescent Medicine Trials Network for HIV/
AIDS Interventions. A clinic-based motivational intervention improves condom use among subgroups of youth living
with HIV-a multicentre randomized controlled trial. J Adolesc Heal. 2011; 49: 193–198.
https://doi.org/10.1016/j.jadohealth.2010.11.252 PMID: 21783053
35. Chandwani S, Abramowitz S, Koenig LJ, Barnes W, D’Angelo L. A multimodal behavioral intervention
to impact adherence and risk behavior among perinatally and behaviorally HIV-infected youth: descrip- tion, delivery,
and receptivity of adolescent impact. AIDS Educ Prev. 2011; 23: 222–35. https://doi.org/ 10.1521/aeap.2011.23.3.222
PMID: 21696241 36. Lee MB, Leibowitz A, Rotheram-Borus MJ. Cost-effectiveness of a behavioral intervention for
sero-
positive youth. AIDS Educ Prev. 2005; 17: 105–118. https://doi.org/10.1521/aeap.17.3.105.62906 PMID: 15899749
37. Markham CM, Shegog R, Leonard AD, Bui TC, Paul ME. +CLICK: harnessing web-based training to
reduce secondary transmission among HIV-positive youth. AIDS Care Psychol Socio-medical Asp AIDS/HIV. 2009;
21: 622–631. https://doi.org/10.1080/09540120802385637 PMID: 19444671 38. Pretorius L, Gibbs A, Crankshaw T,
Willan S. Interventions targeting sexual and reproductive health
and rights outcomes of young people living with HIV: a comprehensive review of current interventions from
sub-Saharan Africa. Glob Health Action. 2015; 8: 28454. https://doi.org/10.3402/gha.v8.28454 PMID: 26534721 39.
Kennedy CE, Medley AM, Sweat MD, O’Reilly KR. Behavioural interventions for HIV positive preven-
tion in developing countries: a systematic review and meta-analysis. Bull World Health Organ. 2010; 88: 615–623.
https://doi.org/10.2471/BLT.09.068213 PMID: 20680127 40. Snyder K, Wallace M, Duby Z, Aquino LDH, Stafford S,
Hosek S, et al. Preliminary results from Hlan-
ganani (Coming Together): A structured support group for HIV-infected adolescents piloted in Cape Town, South
Africa. Child Youth Serv Rev. 2014; 45: 114–121. https://doi.org/10.1016/j.childyouth. 2014.03.027 41. Lightfoot MA,
Kasirye R, Comulada WS, Rotheram-Borus MJ. Efficacy of a culturally adapted interven-
tion for youth living with HIV in Uganda. Prev Sci. 2007; 8: 271–273. https://doi.org/10.1007/s11121- 007-0074-5
PMID: 17846891

42. Senyonyi RM, Underwood L a, Suarez E, Musisi S, Grande TL. Cognitive behavioral therapy group
intervention for HIV transmission risk behavior in perinatally infected adolescents. Health (Irvine Calif). 2012; 4:
1334–1345. 43. Fair C, Wiener LS, Zadeh S, Albright J, Mellins CA, Mancilla M, et al. Reproductive Health Decision-
Making in Perinatally HIV-Infected Adolescents and Young Adults. Matern Child Health J. 2012; 1–12. 44.
Moher D, Liberati A, Tetzlaff J, Altman DG. Reprint—preferred reporting items for systematic reviews
and meta-analyses: the PRISMA statement. Phys Ther. 2009; 89: 873–880. https://doi.org/10.1136/ bmj.b2535 PMID:
19723669 45. Green S, Higgins P., T. J, Alderson P, Clarke M, Mulrow D C, Oxman D A. Cochrane Handbook:
Cochrane Reviews: Ch 8: Assessing risk of bias in included studies. Cochrane Handbook for: System- atic Reviews
of Interventions. 2011. pp. 3–10. 46. Roberts I, Ker K. Cochrane: the unfinished symphony of research synthesis.
Syst Rev. Systematic
Reviews; 2016; 5: 115. https://doi.org/10.1186/s13643-016-0290-9 PMID: 27416925 47. Viegas EO, Tembe N,
Goncalves E, Macovela E, Augusto O, Ismael N, et al. Incidence of HIV and the
Prevalence of HIV, Hepatitis B and Syphilis among Youths in Maputo, Mozambique: A Cohort Study. PLoS ONE.
2015. p. e0121452. https://doi.org/10.1371/journal.pone.0121452 PMID: 25798607 48. Murray J, Farrington DP,
Eisner MP. Drawing conclusions about causes from systematic reviews of
risk factors: The Cambridge Quality Checklists. J Exp Criminol. 2009; 5: 1–23. https://doi.org/10.1007/
s11292-008-9066-0 49. Pantelic M, Shenderovich Y, Cluver LD, Boyes ME. Predictors of internalised HIV-related
stigma: a
systematic review of studies in sub-Saharan Africa. Health Psychol Rev. 2015; 9: 469–490. https://doi.
org/10.1080/17437199.2014.996243 PMID: 25559431 50. Ankunda R, Atuyambe LM, Kiwanuka N. Sexual risk
related behaviour among youth living with HIV in
central Uganda: implications for HIV prevention. Pan Afr Med J. 2016; 24: 1–8. 51. Birungi H, Mugisha JF,
Obare F, Nyombi JK. Sexual behavior and desires among adolescents perinatally infected with human
immunodeficiency virus in Uganda: implications for programming. J Adolesc Heal. 2009; 44: 184–187.
https://doi.org/10.1016/j.jadohealth.2008.06.004 PMID: 19167668 52. Mbalinda SN, Kiwanuka N, Eriksson LE,
Wanyenze RK, Kaye DK. Correlates of ever had sex among
perinatally HIV-infected adolescents in Uganda. Reprod Health. Reproductive Health; 2015; 12: 96.
https://doi.org/10.1186/s12978-015-0082-z PMID: 26475268
53. Obare F, Van Der Kwaak A, Adieri B, Owuor D, Okoth S, Musyoki S, et al. HIV-positive adolescents in
Kenya: Access to Sexual and Reproductive Health Services. KIT Development Policy & Practice. Amsterdam: KIT
Publishers; 2010. pp. 1–54. 54. Toska E, Cluver LD, Hodes RJ, Kidia KK. Sex and secrecy: How HIV-status
disclosure affects safe
sex among HIV-positive adolescents. AIDS Care. 2015; 27: 47–58. https://doi.org/10.1080/09540121. 2015.1071775
PMID: 26616125 55. No ̈stlinger C, Bakeera-Kitaka S, Buyze J, Loos J, Buve ́ A. Factors influencing social
self-disclosure
among adolescents living with HIV in Eastern Africa. AIDS Care. 2015; 27 Suppl 1: 36–46. https://doi.
org/10.1080/09540121.2015.1051501 PMID: 26616124 56. Katusiime C, Kambugu A. Hepatitis B virus infection in
adolescents and young adults with human
immunodeficiency virus infection in an urban clinic in a resource-limited setting. J Int AIDS Soc. 2012; 15: 18433.
https://doi.org/10.7448/IAS.15.6.18433 Poster 57. Banura C, Franceschi S, Doorn van L-J, Arslan A,
Wabwire-Mangen F, Mbidde EK, et al. Infection with human papillomavirus and HIV among young women in
Kampala, Uganda. J Infect Dis. 2008; 197: 555–562. https://doi.org/10.1086/526792 PMID: 18237268 58.
Bakeera-Kitaka S, Nabukeera-Barungi N, No ̈stlinger C, Addy K, Colebunders R. Sexual risk reduction
needs of adolescents living with HIV in a clinical care setting. AIDS Care. 2008; 20: 426–33. https://
doi.org/10.1080/09540120701867099 PMID: 18449819 59. Baryamutuma R, Nabaggala R, Muhairwe LB, Baingana
F. Factors influencing sexual behaviours
among adolescents living with HIV and AIDS in Uganda. 18th International AIDS Conference— AIDS2010. 2010. 60.
Beyeza-Kashesya J, Kaharuza F, Ekstrom AM, Neema S, Kulane A, Mirembe F, et al. To use or not to use a
condom: a prospective cohort study comparing contraceptive practices among HIV-infected and HIV-negative youth
in Uganda. BMC Infect Dis. 2011; 11: 1–11. 61. Birungi H, Obare F, Van Der Kwaak A, Namwebya JH. Maternal
Health Care Utilization Among HIV- Positive Female Adolescents in Kenya. Int Perspect Sex Reprod Health. 2011;
37: 143–149. https:// doi.org/10.1363/3714311 PMID: 21988790
62. Hoffman IF, Martinson FEA, Powers KA, Chilongozi DA, Msiska ED, Kachipapa EI, et al. The year-
long effect of HIV-positive test results on pregnancy intentions, contraceptive use, and pregnancy

incidence among Malawian women. JAIDS J Acquir Immune Defic Syndr. 2008; 47: 477–483. https://
doi.org/10.1097/QAI.0b013e318165dc52 PMID: 18209677 63. Holub CK. Parental/caregiver influence on sexual risk
behaviors among HIV-positive young people in
Kinshasa, the Democratic Republic of the Congo. Dissertation Abstracts International: Section B: The Sciences and
Engineering. 2010. p. 648. 64. Kaggwa E, Hindin M. Psychological well being and sexual behaviour among
HIV-positive young peo-
ple: a case study from Uganda. XIX International AIDS Conference. Washington D.C., USA; 2012. p. WEPE387. 65.
Muyindike W, Fatch R, Steinfield R, Matthews LT, Musinguzi N, Emenyonu NI, et al. Contraceptive
use and associated factors among women enrolling into HIV care in southwestern Uganda. Infect Dis Obstet
Gynecol. 2012; 340782. https://doi.org/10.1155/2012/340782 PMID: 23082069 66. Wanyenze RK, Tumwesigye NM,
Kindyomunda R, Beyeza-Kashesya J, Atuyambe L, Kansiime A,
et al. Uptake of family planning methods and unplanned pregnancies among HIV-infected individuals: A
cross-sectional survey among clients at HIV clinics in Uganda. J Int AIDS Soc. 2011; 14: 35. https://
doi.org/10.1186/1758-2652-14-35 PMID: 21718524 67. Malaju MT, Asale GA. Association of Khat and alcohol use
with HIV infection and age at first sexual ini-
tiation among youths visiting HIV testing and counseling centers in Gamo-Gofa Zone, South West Ethiopia. BMC Int
Health Hum Rights. 2013; 13. https://doi.org/10.1186/1472-698X-13-10 PMID: 23375131 68. Gavin L, Galavotti C,
Dube H, McNaghten AD, Murwirwa M, Khan R, et al. Factors associated with
HIV infection in adolescent females in Zimbabwe. Journal of Adolescent Health. 2006. pp. e11–e18.
https://doi.org/10.1016/j.jadohealth.2006.03.002 PMID: 16982397
69. Gray RH, Wawer MJ, Serwadda D, Sewankambo N, Li C, Wabwire-Mangen F, et al. Population-
based study of fertility in women with HIV-1 infection in Uganda. Lancet. 1998; 351: 98–103. https://
doi.org/10.1016/S0140-6736(97)09381-1 PMID: 9439494 70. Heffron R, Were E, Celum C, Mugo N, Ngure K, Kiarie
J, et al. A prospective study of contraceptive
use among African women in HIV-1 serodiscordant partnerships. Sex Transm Dis. 2010; 37: 621–628.
https://doi.org/10.1097/OLQ.0b013e3181e1a162 PMID: 20601930 71. Hendriksen ES, Pettifor A, Lee S-J, Coates TJ,
Rees H V. Predictors of condom use among young
adults in South Africa: the Reproductive Health and HIV Research Unit National Youth Survey. Am J Public Health.
2007; 97: 1241–1248. https://doi.org/10.2105/AJPH.2006.086009 PMID: 17538062
72. Kembo J. Risk factors associated with HIV infection among young persons aged 15–24 years: evidence from an
in-depth analysis of the 2005–06 Zimbabwe Demographic and Health Survey. SAHARA J. 2012; 9: 54–63.
https://doi.org/10.1080/17290376.2012.683579 PMID: 23237040 73. Pascoe SJS, Langhaug LF, Mavhu W,
Hargreaves JR, Jaffar S, Hayes RJ, et al. Poverty, food insuffi-
ciency and HIV infection and sexual behaviour among young rural Zimbabwean women. PLoS One. 2015; 10:
e0115290–e0115290. https://doi.org/10.1371/journal.pone.0115290 PMID: 25625868 74. Birungi H, Obare F,
Mugisha JF, Evelia H, Nyombi J. Preventive service needs of young people peri-
natally infected with HIV in Uganda. AIDS Care Psychol Socio-medical Asp AIDS/HIV. 2009; 21: 725– 731.
https://doi.org/10.1080/09540120802511901 PMID: 19806488
75. Test FS, Mehta SD, Handler A, Mutimura E, Bamukunde AM, Cohen M. Gender inequities in sexual
risks among youth with HIV in Kigali, Rwanda. Int J STD AIDS. 2012; 23: 394–399. https://doi.org/10.
1258/ijsa.2011.011339 PMID: 22807531 76. Mhalu A, Leyna GH, Mmbaga EJ. Risky behaviours among young
people living with HIV attending
care and treatment clinics in Dar Es Salaam, Tanzania: Implications for prevention with a positive approach. J Int
AIDS Soc. 2013; 16: 1–7. 77. Ankunda R, Atuyambe L, Kiwanuka N. Abstinence and consistent condom use among
youth living
with HIV in urban Uganda: implications for positive prevention. International AIDS Conference— IAS2011. Rome,
Italy; 2011.
78. Nhamo D. Shaping the health of adolescents in zimbabwe (SHAZ!) Key findings and recommenda-
tions from an Economic and Lifeskills intervention addressing SRH and HIV issues among female adolescents
[Internet]. 2014. http://www.avac.org/sites/default/files/event_files/SHAZ!webinarslides.pdf 79. Davidoff-Gore A, Luke
N, Wawire S. Dimensions of poverty and inconsistent condom use among
youth in urban Kenya. AIDS Care. 2011; 23: 1282–1290. https://doi.org/10.1080/09540121.2011. 555744 PMID:
21562992 80. Asante KO, Doku PN. Cultural adaptation of the condom use self efficacy scale (CUSES) in Ghana.
BMC Public Health. 2010; 10: 1–7.

81. Pettifor A, O’Brien K, Macphail C, Miller WC, Rees H. Early coital debut and associated HIV risk factors
among young women and men in South Africa. Int Perspect Sex Reprod Health. 2009; 35: 82–90.
https://doi.org/10.1363/ifpp.35.082.09 PMID: 19620092 82. Toska E, Cluver LD, Boyes ME, Pantelic M, Kuo C. From
“sugar daddies” to “sugar babies”: Quantita-
tively testing the pathway between inequitable sexual relationships, condom use, and adolescent pregnancy in South
Africa. Sex Health. 2015; 12: 59–66. https://doi.org/10.1071/SH14089 PMID: 25702156 83. Steffenson AE, Pettifor
AE, Seage GR 3rd, Rees H V, Cleary PD. Concurrent sexual partnerships and human immunodeficiency virus risk
among South African youth. Sex Transm Dis. United States; 2011; 38: 459–466.
https://doi.org/10.1097/OLQ.0b013e3182080860 PMID: 21258268 84. Bronfenbrenner U. The ecology of human
development: Experiments by nature and design. Harvard
University Press. Cambridge, MA: Harvard University Press; 1979. 85. Mburu G, Ram M, Oxenham D,
Haamujompa C, Iorpenda K, Ferguson L. Responding to adolescents living with HIV in Zambia: A social–ecological
approach. Child Youth Serv Rev. 2014; 45: 9–17. https://doi.org/10.1016/j.childyouth.2014.03.033 86. Stokols D.
Translating social ecological theory into guidelines for community health promotion. Ameri-
can Journal of Health Promotion. 1996. pp. 282–298. https://doi.org/10.4278/0890-1171-10.4.282 PMID: 10159709
87. Nhamo D, Mudekunye-Mahaka I, Chang O, Chingono A, Kadzirange G, Dunbar MS. Factors associ-
ated with gender-based violence and unintended pregnancy among adolescent women living with HIV in Zimbabwe.
International AIDS Conference—IAS2013. 2013. 88. Mellins CA, Nestadt D, Bhana A, Petersen I, Abrams EJ, Alicea
S, et al. Adapting Evidence-Based
Interventions to Meet the Needs of Adolescents Growing Up with HIV in South Africa: The VUKA Case Example.
Glob Soc Welf. 2014; 1: 97–110. https://doi.org/10.1007/s40609-014-0023-8 PMID: 25984440 89. Bhana A, Mellins
CA, Petersen I, Alicea S, Myeza N, Holst H, et al. The VUKA family program: piloting
a family-based psychosocial intervention to promote health and mental health among HIV infected early adolescents
in South Africa. AIDS Care. 2013; 26: 1–11. https://doi.org/10.1080/09540121. 2013.806770 PMID: 23767772 90.
Njika G, Obong’o C, Vandenhoudt H, Adipo D, Murungi I, Loos J, et al. Positive living for a brighter
future: adaptation of a sexual and reproductive health intervention for adolescents living with HIV in Kenya and
Uganda. 18th International AIDS Conference—AIDS2010. Vienna, Austria; 2010. 91. Mupambireyi Z, Bernays S,
Bwakura-Dangarembizi M, Cowan FM. “I don’t feel shy because I will be
among others who are just like me...”: The role of support groups for children perinatally infected with HIV in
Zimbabwe. Children and Youth Services Review. 2014. pp. 106–113. https://doi.org/10.1016/j.
childyouth.2014.03.026 PMID: 25284920 92. Parker L, Maman S, Pettifor AE, Chalachala JL, Edmonds A, Golin CE,
et al. Adaptation of a U.S. evidence-based Positive Prevention intervention for youth living with HIV/AIDS in
Kinshasa, Democratic Republic of the Congo. Eval Program Plann. 2013; 36: 124–135. https://doi.org/10.1016/j.
evalprogplan.2012.09.002 PMID: 23063699 93. No ̈stlinger C, Loos J, Bakeera-Kitaka S, Obong’o CO, Eric W, Buve
A. Translating primary into “posi-
tive” prevention for adolescents in Eastern Africa. Health Promot Int. 2015; 1–12. 94. Hodgson I, Ross J,
Haamujompa C, Gitau-Mburu D. Living as an adolescents with HIV in Zambia—
lived experiences, sexual health and reproductive needs. AIDS Care Psychol Socio-medical Asp AIDS/HIV. 2012; 24:
1204–1210. 95. Buhi ER, Goodson P. Predictors of adolescent sexual behavior and intention: a theory-guided
system-
atic review. J Adolesc Heal. 2007; 40: 4–21. https://doi.org/10.1016/j.jadohealth.2006.09.027 PMID: 17185201 96.
Agha S. A Quasi-Experimental to Assess Study the Impact of Four Adolescent Sexual Health Interven-
tions in Sub-Saharan Africa. Int Fam Plan Perspect. 2002; 28: 67–70–118. 97. Shadish WR, Cook TD,
Campbell DT. Experimental and Quasi-Experimental Designs for Generalised
Causal Inference. Boston, MA, US: Houghton Mifflin Company; 2002. 98. Soeters H, Hatane L, Gittings L, Mark
D. One-stop adolescent shop—Delivering adolescent-friendly
sexual and reproductive health, HIV and TB services in Kafue, Zambia. Paediatric AIDS Treatment for Africa (PATA);
2015. 99. Macleod CI, Tracey T. A decade later: follow-up review of South African research on the conse-
quences of and contributory factors in teen-aged pregnancy. South African J Psychol. 2010; 40: 18– 31.

100. Panday S, Makiwane M, Ranchod C, Letsoalo T. Teenage Pregnancy in South Africa: with a specifi
focus on school-going learners. 101. Makiwane M, Desmond C, Richter L, Udjo E. Is the Child Support Grant
associated with an increase in
teenage fertility in South Africa? Evidence from national surveys and administrative data [Internet]. Methodology.
2006. ww.hsrc.ac.za/Document-2027.phtml 102. Grant MJ, Hallman KK. Pregnancy-related School Dropout and Prior
School Performance in KwaZulu-
Natal, South Africa. Stud Fam Plann. 2008; 39: 369–382. PMID: 19248721 103. Every Woman Every Child.
The Global Strategy for Women’s, Children’s and Adolescents’ Health
(2016–2030). Every Woman Every Child; 2015. 10.1017/CBO9781107415324.004 104. Maynard-tucker G. HIV /
AIDS and family planning services integration: review of prospects for a com-
prehensive approach in sub-Saharan Africa AJAR a comprehensive approach in sub-Saharan Africa. 2011; 37–41.
https://doi.org/10.2989/AJAR.2009.8.4.10.1047 PMID: 25875710 105. Wilkinson L, Moyo F, Henwood R, Runeyi P,
Patel S, Azevedo V De, et al. Youth ART adherence
clubs: Outcomes from an innovative model for HIV positive youth in Khayelitsha, South Africa. 21st International
AIDS Conference—AIDS2016. Durban, South Africa; 2016. http://programme.aids2016.
org/PAGMaterial/eposters/0_7958.pdf 106. Kossow E, Dru ̈phake V, Tolle L, Nkhwalume T, Ntshekisang T, Phoi O.
Teen Mothers Support Group
(TMSG): improving the clinical and psychosocial well being of HIV—positive teenage mothers at the
Botswana—Baylor Children aˆ€TM s Clinical Centre of Excellence (BBCCCOE). 2015; 2015. 107. Pettitt E, Greifinger
R, Phelps R, Bowsky S. Improving adolescent HIV treatment, care, prevention and
family planning services: a multi-country assessment. 19th International AIDS Conference— AIDS2012. Washington
D.C., USA; 2012. 108. Cluver LD, Toska E, Orkin FM, Meinck F, Hodes R, Yakubovich AR, et al. Achieving equity in
HIV-
treatment outcomes: Can social protection improve adolescent ART-adherence in South Africa? AIDS Care. 2016;
28: 73–82. https://doi.org/10.1080/09540121.2016.1179008 PMID: 27392002
109. Toska E, Gittings L, Cluver LD, Hodes RJ, Chademana E, Gutierrez VE. Resourcing resilience: social
protection for HIV prevention amongst children and adolescents in Eastern and Southern Africa. Afri- can J AIDS
Res. 2016; 15: 123–140. https://doi.org/10.2989/16085906.2016.1194299 PMID: 27399042 110. Coates TJ, Richter
L, Caceres C. Behavioural strategies to reduce HIV transmission: how to make
them work better. Lancet. 2008; 372: 669–684. https://doi.org/10.1016/S0140-6736(08)60886-7 PMID: 18687459
111. Jewkes RK, Dunkle K, Nduna M, Shai N. Intimate partner violence, relationship power inequity, and
incidence of HIV infection in young women in South Africa: A cohort study. Lancet. Elsevier Ltd; 2010; 376: 41–48.
https://doi.org/10.1016/S0140-6736(10)60548-X 112. Li Y, Marshall CM, Rees HC, Nunez A, Ezeanolue EE, Ehiri JE.
Intimate partner violence and HIV
infection among women: a systematic review and meta-analysis. J Int AIDS Soc. 2014;17. https://doi.
org/10.7448/IAS.17.1.18845 PMID: 24560342 113. Hoare J, Fouche JP, Spottiswoode B, Donald K, Paul R, Zar H, et
al. A diffusion tensor imaging and
neurocognitive study of HIV positive children who are HAART-naive aˆ€TM slow progressors ‘. 19th International
AIDS Conference—AIDS2012. Washington D.C., USA; 2012. 114. Mbalinda SN, Kiwanuka N, Kaye DK, Eriksson LE.
Reproductive health and lifestyle factors associ-
ated with health-related quality of life among perinatally HIV-infected adolescents in Uganda. Health Qual Life
Outcomes. 2015; 13: 170. https://doi.org/10.1186/s12955-015-0366-6 PMID: 26490047 115. Vale B, Thabeng M.
Redeeming lost mothers: youth antiretroviral treatment and the making of home in
South Africa. Med Anthropol. Routledge; 2016; 1–14. https://doi.org/10.1080/01459740.2016. 1145218 PMID:
26814018
116. Maccarthy S, Rasanathan JJK, Ferguson L, Gruskin S. The pregnancy decisions of HIV-positive
women: the state of knowledge and way forward. Reprod Health Matters. 2012; 20: 119–140. https://
doi.org/10.1016/S0968-8080(12)39641-9 PMID: 23177686 117. Fairlie L, Sipambo N, Fick C, Moultrie H. Focus on
adolescents with HIV and AIDS. South African Med
J. South African Medical Association; 2014; 104: 897. https://doi.org/10.7196/SAMJ.9110
118. Obare F, Birungi H. The limited effect of knowing they are HIV-positive on the sexual and reproductive
experiences and intentions of infected adolescents in Uganda. Popul Stud A J Demogr. 2010; 64: 97– 104.
https://doi.org/10.1080/00324720903427575 PMID: 20087816 119. Fayola T, editor. Teen Life in Africa. Greenwood
Publishing Group; 2004. 120. van Rooyen H, Essack Z, Rochat T, Wight D, Knight L, Bland R, et al. Taking HIV
Testing to Families:
Designing a Family-Based Intervention to Facilitate HIV Testing, Disclosure, and Intergenerational

Communication. Front Public Heal. 2016; 4: 154. https://doi.org/10.3389/fpubh.2016.00154 PMID: 27547750 121.
Chaudhury S, Kirk CM, Ingabire C, Mukunzi S, Nyirandagijimana B, Godfrey K, et al. HIV Status Dis-
closure through Family-Based Intervention Supports Parenting and Child Mental Health in Rwanda. Front Public
Heal. 2016; 4: 138. https://doi.org/10.3389/fpubh.2016.00138 PMID: 27446902 122. Rosebush J, Pettitt E, Offorjebe
A, Boitumelo N, Jibril H, Anabwani G. Teen Talk (Botswana Edition):
a question and answer guide for HIV—positive adolescents. 2015; 2015. 123. Magadi MA, Uchudi J. Onset of
sexual activity among adolescents in HIV/AIDS-affected households
in sub-Saharan Africa. J Biosoc Sci. 2015; 47: 238–257. https://doi.org/10.1017/S0021932014000200 PMID:
24871370 124. Doyle AM, Mavedzenge SN, Plummer ML, Ross DA. The sexual behaviour of adolescents in sub-
Saharan Africa: Patterns and trends from national surveys. Tropical Medicine and International Health. 2012. pp.
796–807. https://doi.org/10.1111/j.1365-3156.2012.03005.x PMID: 22594660 125. Kaggwa EB, Hindin MJ.
Psychological well being and sexual behaviour among HIV-positive young
people: a case study from Uganda. 19th International AIDS Conference—AIDS2012. Washington D. C., USA; 2012.

Toska Et Al. - 2017 - Sex in The Shadow of HIV A Systematic Review of P

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Toska Et Al. - 2017 - Sex in The Shadow of HIV A Systematic Review of P

Diunggah oleh

Hak Cipta:

Format Tersedia

a1111111111 a1111111111 a1111111111 a1111111111 a1111111111

PLOS ONE | https://doi.org/10.1371/journal.pone.0178106 June 5, 2017 1 / 30

RESEARCH ARTICLE Sex in the shadow of HIV: A

Characteristics of included studies

Quality assessment of included studies

Prevalence of sexual risk-taking (Table 3, S6 Table)

Interventions addressing sexual risk-taking among HIV-positive adolescents and youth

Quality of included studies

Implications for future research and practice

Anda mungkin juga menyukai